diff --git a/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-10/lora/README.md b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-10/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-10/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-10/lora/adapter_config.json b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-10/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..ec22fc2bfb7e31cdb0828e6cf23b83bbee872655
--- /dev/null
+++ b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-10/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "qkv_proj",
+    "o_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-10/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-10/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..d7d7ba057c4f043caa5ae9b95bf6c76f7d935058
--- /dev/null
+++ b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-10/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ff89697fb1a253d93be2eefd2926dd29f16d59aab195de38a58b27998cfc866e
+size 37765936
diff --git a/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-100/lora/README.md b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-100/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-100/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-100/lora/adapter_config.json b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-100/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..ec22fc2bfb7e31cdb0828e6cf23b83bbee872655
--- /dev/null
+++ b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-100/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "qkv_proj",
+    "o_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-100/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-100/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..c7a532cb9f432e34ab8a3879e8a1335fe2811320
--- /dev/null
+++ b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-100/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:816900dc9082890a7ab82d7184f8c0e33afb85cb912cd20927365bf2c89b087c
+size 37765936
diff --git a/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-110/lora/README.md b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-110/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-110/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-110/lora/adapter_config.json b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-110/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..ec22fc2bfb7e31cdb0828e6cf23b83bbee872655
--- /dev/null
+++ b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-110/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "qkv_proj",
+    "o_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-110/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-110/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..d600f9b3fbbc4e3506a40340eddee5a199f56deb
--- /dev/null
+++ b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-110/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:95f633833fc088eb24d9dcd5f448d1c31d5f92af2b749d1f5af2946eeb36b420
+size 37765936
diff --git a/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-120/lora/README.md b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-120/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-120/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-120/lora/adapter_config.json b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-120/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..ec22fc2bfb7e31cdb0828e6cf23b83bbee872655
--- /dev/null
+++ b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-120/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "qkv_proj",
+    "o_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-120/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-120/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..0ebba31824a6b9208af04b7009f36e455ea22a91
--- /dev/null
+++ b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-120/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dac655ab839b9bbe96584b63ecc436064038dca14a7297648895d32ceac48e57
+size 37765936
diff --git a/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-130/lora/README.md b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-130/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-130/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-130/lora/adapter_config.json b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-130/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..ec22fc2bfb7e31cdb0828e6cf23b83bbee872655
--- /dev/null
+++ b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-130/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "qkv_proj",
+    "o_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-130/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-130/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..0e5c7d769b6255ed4aac4f63a03ac50d9877503f
--- /dev/null
+++ b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-130/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:12738e458b1b790ea2fe96fdccf27ddc4eb93f05dd16b4f3089b311933b4962b
+size 37765936
diff --git a/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-140/lora/README.md b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-140/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-140/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-140/lora/adapter_config.json b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-140/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..ec22fc2bfb7e31cdb0828e6cf23b83bbee872655
--- /dev/null
+++ b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-140/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "qkv_proj",
+    "o_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-140/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-140/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..3354784c8245cc30e1b3206f8dce38580705520c
--- /dev/null
+++ b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-140/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ffe4fe391824709311f3c72985212c878377df007bf59496d1f14a4309ce7d69
+size 37765936
diff --git a/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-150/lora/README.md b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-150/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-150/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-150/lora/adapter_config.json b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-150/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..ec22fc2bfb7e31cdb0828e6cf23b83bbee872655
--- /dev/null
+++ b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-150/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "qkv_proj",
+    "o_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-150/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-150/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..5dff63296f48c0826bf677e08664185fb608c141
--- /dev/null
+++ b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-150/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6821d7ca10e2afac479bedb7e4a5974baa3559184608ac194c0e8c68f811c26c
+size 37765936
diff --git a/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-160/lora/README.md b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-160/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-160/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-160/lora/adapter_config.json b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-160/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..ec22fc2bfb7e31cdb0828e6cf23b83bbee872655
--- /dev/null
+++ b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-160/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "qkv_proj",
+    "o_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-160/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-160/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..50bef42a8fe22e6f9abe571836da2242dca0fb93
--- /dev/null
+++ b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-160/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a8ac735cb378f1b4ec26af9409254c10a1857074f24a20fa87e95693bf82686a
+size 37765936
diff --git a/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-170/lora/README.md b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-170/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-170/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-170/lora/adapter_config.json b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-170/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..ec22fc2bfb7e31cdb0828e6cf23b83bbee872655
--- /dev/null
+++ b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-170/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "qkv_proj",
+    "o_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-170/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-170/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..13daf59bf35951c2aaa0fd822f189865787af291
--- /dev/null
+++ b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-170/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0d780993f07cd0f197554189dde3f93b5577ba887a40f20e474fa64a40f9bef9
+size 37765936
diff --git a/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-180/lora/README.md b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-180/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-180/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-180/lora/adapter_config.json b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-180/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..ec22fc2bfb7e31cdb0828e6cf23b83bbee872655
--- /dev/null
+++ b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-180/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "qkv_proj",
+    "o_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-180/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-180/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..1bb47c51720ff24e3f1346188d4a3193e6a0f8ae
--- /dev/null
+++ b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-180/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:91bb7271e5a6036d0e33549c6ac649bc8f2167335ef7db1a41bfdc4d1ac2c13f
+size 37765936
diff --git a/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-190/lora/README.md b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-190/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-190/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-190/lora/adapter_config.json b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-190/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..ec22fc2bfb7e31cdb0828e6cf23b83bbee872655
--- /dev/null
+++ b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-190/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "qkv_proj",
+    "o_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-190/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-190/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..3efad76c995ce36e2489376ecefa0a1807819a9e
--- /dev/null
+++ b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-190/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cfb3613cb4e0a79b729da8d24a2c806125268d6aa5786f0157e6a637090f40cc
+size 37765936
diff --git a/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-20/lora/README.md b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-20/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-20/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-20/lora/adapter_config.json b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-20/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..ec22fc2bfb7e31cdb0828e6cf23b83bbee872655
--- /dev/null
+++ b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-20/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "qkv_proj",
+    "o_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-20/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-20/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..c837747e823f10114d289ce455047f84a99e1354
--- /dev/null
+++ b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-20/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:26ac184c456fce8f43aaff9d2449b31405ea05737bd3083ff4fd378d09546824
+size 37765936
diff --git a/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-30/lora/README.md b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-30/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-30/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-30/lora/adapter_config.json b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-30/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..ec22fc2bfb7e31cdb0828e6cf23b83bbee872655
--- /dev/null
+++ b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-30/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "qkv_proj",
+    "o_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-30/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-30/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..ff46c85cab6900218e05adce4928f10baae3114f
--- /dev/null
+++ b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-30/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f59c1f2e7e7dd07e0639f6732cfef4cec48b240196f4ee9421cd3c70a2811709
+size 37765936
diff --git a/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-40/lora/README.md b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-40/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-40/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-40/lora/adapter_config.json b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-40/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..ec22fc2bfb7e31cdb0828e6cf23b83bbee872655
--- /dev/null
+++ b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-40/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "qkv_proj",
+    "o_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-40/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-40/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..0320d9d029d3a412990e95dc41b3a2bd2a23bb0f
--- /dev/null
+++ b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-40/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e0f27cb77e7018583a784963dc0934610e7b2ed020223da8f7f60731f03fb8c7
+size 37765936
diff --git a/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-50/lora/README.md b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-50/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-50/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-50/lora/adapter_config.json b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-50/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..ec22fc2bfb7e31cdb0828e6cf23b83bbee872655
--- /dev/null
+++ b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-50/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "qkv_proj",
+    "o_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-50/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-50/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..0b8bc5c817d9331646424fa4b89b5637a527bb43
--- /dev/null
+++ b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-50/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ca7f24fec862fd819b7ae97f5ca4a07adf4ef03d18afa8aa4f7f0285048a25fb
+size 37765936
diff --git a/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-60/lora/README.md b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-60/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-60/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-60/lora/adapter_config.json b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-60/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..ec22fc2bfb7e31cdb0828e6cf23b83bbee872655
--- /dev/null
+++ b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-60/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "qkv_proj",
+    "o_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-60/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-60/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..9fb2413c0055a82c86867e5734beac1f823f9a20
--- /dev/null
+++ b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-60/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fb18a499197982885cfe4b22b2f94b29dac4df97f5ebb046db3cb31f5640fe8e
+size 37765936
diff --git a/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-70/lora/README.md b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-70/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-70/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-70/lora/adapter_config.json b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-70/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..ec22fc2bfb7e31cdb0828e6cf23b83bbee872655
--- /dev/null
+++ b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-70/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "qkv_proj",
+    "o_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-70/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-70/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..f897af009d2ff778b1bb81ef65d64bb1e6afb75f
--- /dev/null
+++ b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-70/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c8d123c846b1044555d18e72ab8715c5d6d70dd24227ce8ac04ca911aba29f57
+size 37765936
diff --git a/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-80/lora/README.md b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-80/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-80/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-80/lora/adapter_config.json b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-80/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..ec22fc2bfb7e31cdb0828e6cf23b83bbee872655
--- /dev/null
+++ b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-80/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "qkv_proj",
+    "o_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-80/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-80/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..1038090c4998a0f6c50def70230e24c1690e8ffa
--- /dev/null
+++ b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-80/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:73e1a3741548b55c9e22af0b6eeaab574191ac071910b74974f1f29de5fb99df
+size 37765936
diff --git a/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-90/lora/README.md b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-90/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-90/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-90/lora/adapter_config.json b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-90/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..ec22fc2bfb7e31cdb0828e6cf23b83bbee872655
--- /dev/null
+++ b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-90/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "qkv_proj",
+    "o_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-90/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-90/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..3d321fc54f74446dec853d23fcfeb52da613e2d4
--- /dev/null
+++ b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/checkpoints/checkpoint-90/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:078147d55688c68be348cf0ea46bc4646035f4b4b17752646f23184194d92937
+size 37765936
diff --git a/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/environment_summary.txt b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/environment_summary.txt
new file mode 100644
index 0000000000000000000000000000000000000000..841a4dc1241c74176f91fb0ffb931761a2feebfe
--- /dev/null
+++ b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/environment_summary.txt
@@ -0,0 +1,21 @@
+{
+  "environment": {
+    "CUDA_VISIBLE_DEVICES": "4,5,6,7",
+    "HF_HOME": "/home/wenting/.cache/huggingface",
+    "HF_HUB_CACHE": "/home/wenting/.cache/huggingface/hub",
+    "OMNIGEN_CODE_ROOT": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint",
+    "OUTPUT_DIR": "outputs/fix_win/statcm_win1_10full_n001_bs64_steps200",
+    "REWARD_MASK_CHANNELS": "0,1",
+    "SAVE_DIR": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/flow_grpo_cxr/logs/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200",
+    "SFT_LORA_PATH": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint/results_new/scratch_15k",
+    "TEST_JSONL": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_test.jsonl",
+    "TRAIN_JSONL": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_train.jsonl",
+    "WANDB_MODE": "online",
+    "WANDB_NAME": "statcm_win1_10full_n001_bs64_h20",
+    "WANDB_PROJECT": "fixwin_h20"
+  },
+  "platform": "Linux-5.4.241-1-tlinux4-0017.7-x86_64-with-glibc2.38",
+  "python": "3.10.20 (main, Mar 11 2026, 17:46:40) [GCC 14.3.0]",
+  "repo_root": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/flow_grpo_cxr",
+  "timestamp": "2026-06-08T22:39:08"
+}
diff --git a/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/git_head.txt b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/git_head.txt
new file mode 100644
index 0000000000000000000000000000000000000000..910c3da6325e61af26279bd74233576c2a0530b2
--- /dev/null
+++ b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/git_head.txt
@@ -0,0 +1 @@
+5c0a81d87d64ffd6ca4e4428794155a290835637
diff --git a/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/git_status.txt b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/git_status.txt
new file mode 100644
index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391
diff --git a/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/launch_config.txt b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/launch_config.txt
new file mode 100644
index 0000000000000000000000000000000000000000..0e7347f6fcb6696056f38cad64ef7ca1514d4c93
--- /dev/null
+++ b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/launch_config.txt
@@ -0,0 +1,49 @@
+timestamp=20260608_223611
+repo_path=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/flow_grpo_cxr
+config_entry=config/grpo.py:general_radiomics_omnigen_4gpu_kl
+server_profile=
+omnigen_code_root=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint
+sft_lora_path=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint/results_new/scratch_15k
+dataset_root=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset
+train_jsonl=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_train.jsonl
+test_jsonl=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_test.jsonl
+dataset_path_remap_from=/home/wenting/dataset/LIDC_IDRI
+dataset_path_remap_to=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset
+mirp_repo_root=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/mirp
+save_dir=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/flow_grpo_cxr/logs/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200
+output_dir=outputs/fix_win/statcm_win1_10full_n001_bs64_steps200
+num_steps=10
+eval_num_steps=50
+noise_level=0.01
+sde_type=cps
+sde_window_enabled=1
+sde_window_start=1
+sde_window_end=10
+sde_window_mode=fixed
+sde_window_count=9
+sde_window_train_only=1
+sde_window_logprob_only=0
+SDE_WINDOW_ENABLED=1
+SDE_WINDOW_START=1
+SDE_WINDOW_END=10
+SDE_WINDOW_MODE=fixed
+SDE_WINDOW_COUNT=9
+guidance_scale=2.5
+img_guidance_scale=2.0
+num_image_per_prompt=4
+train_batch_size=64
+test_batch_size=32
+num_batches_per_epoch=2
+kl_beta=0.005
+learning_rate=1e-4
+max_train_steps=200
+reward_feature_prefixes=stat_,cm_
+reward_base_feature_families=stat,cm
+reward_mask_channels=0,1
+num_processes=4
+wandb_project=fixwin_h20
+wandb_group=manual_img_only
+wandb_name=statcm_win1_10full_n001_bs64_h20
+wandb_tags=img-only,scratch30k,manual,noise0.01,bs64
+wandb_mode=online
+wandb_resume=never
diff --git a/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/resolved_config.json b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/resolved_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..5a55999a2353bd9c6aac0b862468b745301bc786
--- /dev/null
+++ b/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200/resolved_config.json
@@ -0,0 +1,164 @@
+{
+  "activation_checkpointing": true,
+  "allow_tf32": true,
+  "config_entry": "config/grpo.py:general_radiomics_omnigen_4gpu_kl",
+  "dataset": {
+    "dataset_path_remap_from": "/home/wenting/dataset/LIDC_IDRI",
+    "dataset_path_remap_to": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset",
+    "path_replacements": [
+      [
+        "/home/wenting/dataset/LIDC_IDRI",
+        "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset"
+      ]
+    ],
+    "root": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset",
+    "test_jsonl": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_test.jsonl",
+    "train_jsonl": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_train.jsonl"
+  },
+  "debug": false,
+  "eval_freq": 30,
+  "fsdp_optimizer_offload": false,
+  "logdir": "logs",
+  "mixed_precision": "bf16",
+  "num_checkpoint_limit": 5,
+  "num_epochs": 100000,
+  "per_prompt_stat_tracking": true,
+  "pretrained": {
+    "local_code_root": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint",
+    "model": "Shitao/OmniGen-v1",
+    "revision": "main",
+    "vae_path": null
+  },
+  "prompt_fn": "radiomics_edit",
+  "prompt_fn_kwargs": {},
+  "resolution": 256,
+  "resume_from": null,
+  "reward_fn": {
+    "radiomics": {
+      "base_feature_families": [
+        "stat",
+        "cm"
+      ],
+      "data_root": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset",
+      "dataset_path_remap_from": "/home/wenting/dataset/LIDC_IDRI",
+      "dataset_path_remap_to": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset",
+      "distance_metric": "relative_l1",
+      "feature_prefixes": [
+        "stat_",
+        "cm_"
+      ],
+      "fixed_bin_number": 32,
+      "gt_root": null,
+      "mask_channel": 0,
+      "mask_channels": [
+        0,
+        1
+      ],
+      "mask_root": null,
+      "mirp_repo_root": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/mirp",
+      "missing_score": -10.0,
+      "pixel_spacing_mm": 1.0,
+      "relative_error_clip": 10.0,
+      "reward_scale": 1.0,
+      "roi_name": "lung_roi",
+      "weight": 1.0
+    }
+  },
+  "run_name": "2026.06.08_22.39.04",
+  "sample": {
+    "batch_size": 8,
+    "eval_guidance_scale": 2.5,
+    "eval_img_guidance_scale": 2.0,
+    "eval_num_steps": 50,
+    "global_std": true,
+    "guidance_scale": 2.5,
+    "img_guidance_scale": 2.0,
+    "max_eval_batches": 2,
+    "max_input_image_size": 256,
+    "noise_level": 0.01,
+    "num_batches_per_epoch": 2,
+    "num_image_per_prompt": 4,
+    "num_steps": 10,
+    "same_latent": false,
+    "sde_type": "cps",
+    "sde_window_enabled": true,
+    "sde_window_end": 10,
+    "sde_window_logprob_only": false,
+    "sde_window_mode": "fixed",
+    "sde_window_range": [
+      0,
+      10
+    ],
+    "sde_window_size": 2,
+    "sde_window_start": 1,
+    "sde_window_train_only": true,
+    "test_batch_size": 32,
+    "train_batch_size": 64,
+    "use_img_guidance": true,
+    "use_input_image_size_as_output": false
+  },
+  "save_dir": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/flow_grpo_cxr/logs/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200",
+  "save_freq": 5,
+  "seed": 42,
+  "stage2_extra": {
+    "base_model": "Shitao/OmniGen-v1",
+    "mask_modules_path": null,
+    "omnigen_code_root": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint",
+    "reward_mask_channels": "0,1",
+    "rl_lora_target_modules": [
+      "qkv_proj",
+      "o_proj"
+    ],
+    "save_dir": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/flow_grpo_cxr/logs/radiomics/fix_win/statcm_win1_10full_n001_bs64_steps200",
+    "sde_window_enabled": true,
+    "sde_window_end": 10,
+    "sde_window_mode": "fixed",
+    "sde_window_start": 1,
+    "sft_lora_path": null,
+    "test_jsonl": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_test.jsonl",
+    "train_jsonl": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_train.jsonl",
+    "train_timestep_indices": [
+      1,
+      2,
+      3,
+      4,
+      5,
+      6,
+      7,
+      8,
+      9
+    ]
+  },
+  "train": {
+    "adam_beta1": 0.9,
+    "adam_beta2": 0.999,
+    "adam_epsilon": 1e-08,
+    "adam_weight_decay": 0.0001,
+    "adv_clip_max": 5,
+    "batch_size": 64,
+    "beta": 0.005,
+    "cfg": true,
+    "clip_range": 0.0001,
+    "clip_range_gt": 0.0001,
+    "clip_range_lt": 0.0001,
+    "condition_dropout_prob": 0.01,
+    "ema": false,
+    "gradient_accumulation_steps": 1,
+    "learning_rate": 0.0001,
+    "lora_alpha": 64,
+    "lora_path": null,
+    "lora_rank": 32,
+    "max_grad_norm": 1.0,
+    "max_train_steps": 200,
+    "merge_lora_path": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint/results_new/scratch_15k",
+    "num_inner_epochs": 1,
+    "rl_lora_target_modules": [
+      "qkv_proj",
+      "o_proj"
+    ],
+    "timestep_fraction": 1.0,
+    "timestep_shift": 3.0,
+    "use_8bit_adam": false
+  },
+  "use_lora": true
+}
diff --git a/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-10/lora/README.md b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-10/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-10/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-10/lora/adapter_config.json b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-10/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..5caca2d3fe4f842c2d173e65caa2fb5f52a00540
--- /dev/null
+++ b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-10/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "qkv_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-10/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-10/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..e60e0950d87703be23a80af14ec68f3c4bc70ffd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-10/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a32e09ab98b5ef96dfe3873e2d8289759c2064d00d9ec0045ef962e1631bdd36
+size 37765936
diff --git a/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-100/lora/README.md b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-100/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-100/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-100/lora/adapter_config.json b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-100/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..5caca2d3fe4f842c2d173e65caa2fb5f52a00540
--- /dev/null
+++ b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-100/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "qkv_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-100/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-100/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..d4fac859d98b315f1299ba3bc70ce4d65347ca6d
--- /dev/null
+++ b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-100/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dd0fa6d7ac2d07d0d2e6dc0876ad31810608473a663ddbc3ef72324d61a1c3bc
+size 37765936
diff --git a/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-110/lora/README.md b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-110/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-110/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-110/lora/adapter_config.json b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-110/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..5caca2d3fe4f842c2d173e65caa2fb5f52a00540
--- /dev/null
+++ b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-110/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "qkv_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-110/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-110/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..664cbce56dece9eec176215a638ded1966e8dfb2
--- /dev/null
+++ b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-110/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3538a806bd2f322e435295f84ebf8fce0f5ac4adae1e387b2377cd975318ce87
+size 37765936
diff --git a/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-120/lora/README.md b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-120/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-120/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-120/lora/adapter_config.json b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-120/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..5caca2d3fe4f842c2d173e65caa2fb5f52a00540
--- /dev/null
+++ b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-120/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "qkv_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-120/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-120/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..f84a8e7970bb647fdacd77393b97431476d51bab
--- /dev/null
+++ b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-120/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8e773baf28195ec30e27595d46db0db8f2aa38e2f34e9efbc90741f4b7f3db7d
+size 37765936
diff --git a/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-130/lora/README.md b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-130/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-130/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-130/lora/adapter_config.json b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-130/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..5caca2d3fe4f842c2d173e65caa2fb5f52a00540
--- /dev/null
+++ b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-130/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "qkv_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-130/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-130/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..2ce77b0324b8e93dd72a93f44fd2f65424942c63
--- /dev/null
+++ b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-130/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1d68568439a1c8d03561e9a475b3df65272387596d2695f2f38443d4000c1331
+size 37765936
diff --git a/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-140/lora/README.md b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-140/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-140/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-140/lora/adapter_config.json b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-140/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..5caca2d3fe4f842c2d173e65caa2fb5f52a00540
--- /dev/null
+++ b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-140/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "qkv_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-140/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-140/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..d98b3734d07cdd10f35d33833f4f8eabdbf1c57d
--- /dev/null
+++ b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-140/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e779495584f232d199eaf8438a687e7c7d0ff4df8d0f4509ec22c31d5169271d
+size 37765936
diff --git a/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-150/lora/README.md b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-150/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-150/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-150/lora/adapter_config.json b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-150/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..5caca2d3fe4f842c2d173e65caa2fb5f52a00540
--- /dev/null
+++ b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-150/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "qkv_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-150/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-150/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..7b34489fff8024d45ec96415d94a3a1c1730fef2
--- /dev/null
+++ b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-150/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a59cbc706940484b3780023c5938f9181cfabcb34c47da085fd1893e2db86aa6
+size 37765936
diff --git a/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-160/lora/README.md b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-160/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-160/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-160/lora/adapter_config.json b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-160/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..5caca2d3fe4f842c2d173e65caa2fb5f52a00540
--- /dev/null
+++ b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-160/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "qkv_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-160/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-160/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..cf0f7051ce349fad3402f29597a66155c2443af6
--- /dev/null
+++ b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-160/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cda27a8f0e3996f49066a466672e50c17f2a1ee953fcfac9ed09c5514995f379
+size 37765936
diff --git a/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-170/lora/README.md b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-170/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-170/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-170/lora/adapter_config.json b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-170/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..5caca2d3fe4f842c2d173e65caa2fb5f52a00540
--- /dev/null
+++ b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-170/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "qkv_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-170/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-170/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..103e466f5864bd1bf61601c7f5060fbaf8b6a47d
--- /dev/null
+++ b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-170/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:76bc6be86cb1caf00d7a6065c5d59031e591bb113f6ceb9f1a17873e02069e64
+size 37765936
diff --git a/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-180/lora/README.md b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-180/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-180/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-180/lora/adapter_config.json b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-180/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..5caca2d3fe4f842c2d173e65caa2fb5f52a00540
--- /dev/null
+++ b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-180/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "qkv_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-180/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-180/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..0e1eac9f7f33b1c45646832ae023eb8957ebc864
--- /dev/null
+++ b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-180/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:426fc2bf06063406bab4c2fe60d6b057e78338c359b97d810d3afe6825d2bdbe
+size 37765936
diff --git a/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-190/lora/README.md b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-190/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-190/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-190/lora/adapter_config.json b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-190/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..5caca2d3fe4f842c2d173e65caa2fb5f52a00540
--- /dev/null
+++ b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-190/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "qkv_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-190/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-190/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..4902bf4547857f5f0045b00e28490ed9938cff48
--- /dev/null
+++ b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-190/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1f61c59e63f83f3a7cfbafa72705fa688b02a4e5a8f17286f9924e8199098a93
+size 37765936
diff --git a/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-20/lora/README.md b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-20/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-20/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-20/lora/adapter_config.json b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-20/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..5caca2d3fe4f842c2d173e65caa2fb5f52a00540
--- /dev/null
+++ b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-20/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "qkv_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-20/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-20/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..ef570968b2e1b2b116d3c48bd5bf0af8cd701f53
--- /dev/null
+++ b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-20/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1ba4db00e0afc515921a7381c6b1a69f1eb74521a170c5d318c99844788ebb4b
+size 37765936
diff --git a/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-30/lora/README.md b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-30/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-30/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-30/lora/adapter_config.json b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-30/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..5caca2d3fe4f842c2d173e65caa2fb5f52a00540
--- /dev/null
+++ b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-30/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "qkv_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-30/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-30/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..b173c101b9317c3467905266c2abef256f32295a
--- /dev/null
+++ b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-30/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:acda441c9f4eaf85da0a66ca1bd3bf304326f5c5003e61a6a0b79f8999ca7ddc
+size 37765936
diff --git a/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-40/lora/README.md b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-40/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-40/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-40/lora/adapter_config.json b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-40/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..5caca2d3fe4f842c2d173e65caa2fb5f52a00540
--- /dev/null
+++ b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-40/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "qkv_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-40/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-40/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..e2175327fc91e8429a9d67a1c8b0b68bc0d06175
--- /dev/null
+++ b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-40/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cdb7e171c935614e7494cd8daa6f720e218fca5167839d2b1b90706596d4610b
+size 37765936
diff --git a/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-50/lora/README.md b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-50/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-50/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-50/lora/adapter_config.json b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-50/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..5caca2d3fe4f842c2d173e65caa2fb5f52a00540
--- /dev/null
+++ b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-50/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "qkv_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-50/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-50/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..fbaadd17fcab1cb8a514a7796cc871fd46a83fa2
--- /dev/null
+++ b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-50/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:30317ce1355084ad8f3a26f726c67e5f0bdfef02ff62b1987b17dee8bf26ffb6
+size 37765936
diff --git a/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-60/lora/README.md b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-60/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-60/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-60/lora/adapter_config.json b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-60/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..5caca2d3fe4f842c2d173e65caa2fb5f52a00540
--- /dev/null
+++ b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-60/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "qkv_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-60/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-60/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..90ee9fd3d364a212354a7db107a84f766b762cea
--- /dev/null
+++ b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-60/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6cc3182571be978f5fab4e4f27a6e911a7792aeed9eff80aa013d375a443a559
+size 37765936
diff --git a/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-70/lora/README.md b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-70/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-70/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-70/lora/adapter_config.json b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-70/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..5caca2d3fe4f842c2d173e65caa2fb5f52a00540
--- /dev/null
+++ b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-70/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "qkv_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-70/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-70/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..74720f19e469e4f1120c8b2cca4ce0b12bf369c0
--- /dev/null
+++ b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-70/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:613267bac41669de794c7e53e50588abd2d2076334ea138c96215fa37d32ecd5
+size 37765936
diff --git a/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-80/lora/README.md b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-80/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-80/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-80/lora/adapter_config.json b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-80/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..5caca2d3fe4f842c2d173e65caa2fb5f52a00540
--- /dev/null
+++ b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-80/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "qkv_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-80/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-80/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..3a8d8af10648b8c42b4c47096372280e2afd6265
--- /dev/null
+++ b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-80/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ac7356939f26ceb2abeea0dee90a58061e9ed01f560b5ba4072e1f1f548aef76
+size 37765936
diff --git a/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-90/lora/README.md b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-90/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-90/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-90/lora/adapter_config.json b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-90/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..5caca2d3fe4f842c2d173e65caa2fb5f52a00540
--- /dev/null
+++ b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-90/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "qkv_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-90/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-90/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..b5d3c67e4c13a64a76596dd3090d13c4e14ab5fc
--- /dev/null
+++ b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/checkpoints/checkpoint-90/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d8dfedb1f9319908ccd85cf88be5b0b10f7dbc8d5411fff47e12a8426ef28b60
+size 37765936
diff --git a/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/environment_summary.txt b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/environment_summary.txt
new file mode 100644
index 0000000000000000000000000000000000000000..276561cbcc6984659df3f8b7be75a710f97b08a1
--- /dev/null
+++ b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/environment_summary.txt
@@ -0,0 +1,21 @@
+{
+  "environment": {
+    "CUDA_VISIBLE_DEVICES": "0,1,2,3",
+    "HF_HOME": "/home/wenting/.cache/huggingface",
+    "HF_HUB_CACHE": "/home/wenting/.cache/huggingface/hub",
+    "OMNIGEN_CODE_ROOT": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint",
+    "OUTPUT_DIR": "outputs/fix_win/statcm_win4_6_n001_bs64_steps200",
+    "REWARD_MASK_CHANNELS": "0,1",
+    "SAVE_DIR": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/flow_grpo_cxr/logs/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200",
+    "SFT_LORA_PATH": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint/results_new/scratch_15k",
+    "TEST_JSONL": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_test.jsonl",
+    "TRAIN_JSONL": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_train.jsonl",
+    "WANDB_MODE": "online",
+    "WANDB_NAME": "statcm_win4_6_n001_bs64_h20",
+    "WANDB_PROJECT": "fixwin_h20"
+  },
+  "platform": "Linux-5.4.241-1-tlinux4-0017.7-x86_64-with-glibc2.38",
+  "python": "3.10.20 (main, Mar 11 2026, 17:46:40) [GCC 14.3.0]",
+  "repo_root": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/flow_grpo_cxr",
+  "timestamp": "2026-06-08T22:38:42"
+}
diff --git a/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/git_head.txt b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/git_head.txt
new file mode 100644
index 0000000000000000000000000000000000000000..910c3da6325e61af26279bd74233576c2a0530b2
--- /dev/null
+++ b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/git_head.txt
@@ -0,0 +1 @@
+5c0a81d87d64ffd6ca4e4428794155a290835637
diff --git a/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/git_status.txt b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/git_status.txt
new file mode 100644
index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391
diff --git a/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/launch_config.txt b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/launch_config.txt
new file mode 100644
index 0000000000000000000000000000000000000000..14fca396cd5f02846343550916218bc76e2d818e
--- /dev/null
+++ b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/launch_config.txt
@@ -0,0 +1,49 @@
+timestamp=20260608_223446
+repo_path=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/flow_grpo_cxr
+config_entry=config/grpo.py:general_radiomics_omnigen_4gpu_kl
+server_profile=
+omnigen_code_root=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint
+sft_lora_path=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint/results_new/scratch_15k
+dataset_root=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset
+train_jsonl=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_train.jsonl
+test_jsonl=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_test.jsonl
+dataset_path_remap_from=/home/wenting/dataset/LIDC_IDRI
+dataset_path_remap_to=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset
+mirp_repo_root=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/mirp
+save_dir=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/flow_grpo_cxr/logs/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200
+output_dir=outputs/fix_win/statcm_win4_6_n001_bs64_steps200
+num_steps=10
+eval_num_steps=50
+noise_level=0.01
+sde_type=cps
+sde_window_enabled=1
+sde_window_start=4
+sde_window_end=6
+sde_window_mode=fixed
+sde_window_count=2
+sde_window_train_only=1
+sde_window_logprob_only=0
+SDE_WINDOW_ENABLED=1
+SDE_WINDOW_START=4
+SDE_WINDOW_END=6
+SDE_WINDOW_MODE=fixed
+SDE_WINDOW_COUNT=2
+guidance_scale=2.5
+img_guidance_scale=2.0
+num_image_per_prompt=4
+train_batch_size=64
+test_batch_size=32
+num_batches_per_epoch=2
+kl_beta=0.005
+learning_rate=1e-4
+max_train_steps=200
+reward_feature_prefixes=stat_,cm_
+reward_base_feature_families=stat,cm
+reward_mask_channels=0,1
+num_processes=4
+wandb_project=fixwin_h20
+wandb_group=manual_img_only
+wandb_name=statcm_win4_6_n001_bs64_h20
+wandb_tags=img-only,scratch30k,manual,noise0.01,bs64
+wandb_mode=online
+wandb_resume=never
diff --git a/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/resolved_config.json b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/resolved_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..10279fb1ad87932f7907cd09f9413d2b52184ff3
--- /dev/null
+++ b/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200/resolved_config.json
@@ -0,0 +1,157 @@
+{
+  "activation_checkpointing": true,
+  "allow_tf32": true,
+  "config_entry": "config/grpo.py:general_radiomics_omnigen_4gpu_kl",
+  "dataset": {
+    "dataset_path_remap_from": "/home/wenting/dataset/LIDC_IDRI",
+    "dataset_path_remap_to": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset",
+    "path_replacements": [
+      [
+        "/home/wenting/dataset/LIDC_IDRI",
+        "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset"
+      ]
+    ],
+    "root": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset",
+    "test_jsonl": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_test.jsonl",
+    "train_jsonl": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_train.jsonl"
+  },
+  "debug": false,
+  "eval_freq": 30,
+  "fsdp_optimizer_offload": false,
+  "logdir": "logs",
+  "mixed_precision": "bf16",
+  "num_checkpoint_limit": 5,
+  "num_epochs": 100000,
+  "per_prompt_stat_tracking": true,
+  "pretrained": {
+    "local_code_root": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint",
+    "model": "Shitao/OmniGen-v1",
+    "revision": "main",
+    "vae_path": null
+  },
+  "prompt_fn": "radiomics_edit",
+  "prompt_fn_kwargs": {},
+  "resolution": 256,
+  "resume_from": null,
+  "reward_fn": {
+    "radiomics": {
+      "base_feature_families": [
+        "stat",
+        "cm"
+      ],
+      "data_root": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset",
+      "dataset_path_remap_from": "/home/wenting/dataset/LIDC_IDRI",
+      "dataset_path_remap_to": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset",
+      "distance_metric": "relative_l1",
+      "feature_prefixes": [
+        "stat_",
+        "cm_"
+      ],
+      "fixed_bin_number": 32,
+      "gt_root": null,
+      "mask_channel": 0,
+      "mask_channels": [
+        0,
+        1
+      ],
+      "mask_root": null,
+      "mirp_repo_root": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/mirp",
+      "missing_score": -10.0,
+      "pixel_spacing_mm": 1.0,
+      "relative_error_clip": 10.0,
+      "reward_scale": 1.0,
+      "roi_name": "lung_roi",
+      "weight": 1.0
+    }
+  },
+  "run_name": "2026.06.08_22.38.38",
+  "sample": {
+    "batch_size": 8,
+    "eval_guidance_scale": 2.5,
+    "eval_img_guidance_scale": 2.0,
+    "eval_num_steps": 50,
+    "global_std": true,
+    "guidance_scale": 2.5,
+    "img_guidance_scale": 2.0,
+    "max_eval_batches": 2,
+    "max_input_image_size": 256,
+    "noise_level": 0.01,
+    "num_batches_per_epoch": 2,
+    "num_image_per_prompt": 4,
+    "num_steps": 10,
+    "same_latent": false,
+    "sde_type": "cps",
+    "sde_window_enabled": true,
+    "sde_window_end": 6,
+    "sde_window_logprob_only": false,
+    "sde_window_mode": "fixed",
+    "sde_window_range": [
+      0,
+      10
+    ],
+    "sde_window_size": 2,
+    "sde_window_start": 4,
+    "sde_window_train_only": true,
+    "test_batch_size": 32,
+    "train_batch_size": 64,
+    "use_img_guidance": true,
+    "use_input_image_size_as_output": false
+  },
+  "save_dir": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/flow_grpo_cxr/logs/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200",
+  "save_freq": 5,
+  "seed": 42,
+  "stage2_extra": {
+    "base_model": "Shitao/OmniGen-v1",
+    "mask_modules_path": null,
+    "omnigen_code_root": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint",
+    "reward_mask_channels": "0,1",
+    "rl_lora_target_modules": [
+      "qkv_proj",
+      "o_proj"
+    ],
+    "save_dir": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/flow_grpo_cxr/logs/radiomics/fix_win/statcm_win4_6_n001_bs64_steps200",
+    "sde_window_enabled": true,
+    "sde_window_end": 6,
+    "sde_window_mode": "fixed",
+    "sde_window_start": 4,
+    "sft_lora_path": null,
+    "test_jsonl": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_test.jsonl",
+    "train_jsonl": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_train.jsonl",
+    "train_timestep_indices": [
+      4,
+      5
+    ]
+  },
+  "train": {
+    "adam_beta1": 0.9,
+    "adam_beta2": 0.999,
+    "adam_epsilon": 1e-08,
+    "adam_weight_decay": 0.0001,
+    "adv_clip_max": 5,
+    "batch_size": 64,
+    "beta": 0.005,
+    "cfg": true,
+    "clip_range": 0.0001,
+    "clip_range_gt": 0.0001,
+    "clip_range_lt": 0.0001,
+    "condition_dropout_prob": 0.01,
+    "ema": false,
+    "gradient_accumulation_steps": 1,
+    "learning_rate": 0.0001,
+    "lora_alpha": 64,
+    "lora_path": null,
+    "lora_rank": 32,
+    "max_grad_norm": 1.0,
+    "max_train_steps": 200,
+    "merge_lora_path": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint/results_new/scratch_15k",
+    "num_inner_epochs": 1,
+    "rl_lora_target_modules": [
+      "qkv_proj",
+      "o_proj"
+    ],
+    "timestep_fraction": 1.0,
+    "timestep_shift": 3.0,
+    "use_8bit_adam": false
+  },
+  "use_lora": true
+}
diff --git a/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-10/lora/README.md b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-10/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-10/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-10/lora/adapter_config.json b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-10/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..ec22fc2bfb7e31cdb0828e6cf23b83bbee872655
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-10/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "qkv_proj",
+    "o_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-10/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-10/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..aa52e26109b2ac75d5dbd1eca4252c4356058b33
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-10/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:478ae41560f4a69c35c44ba8c630934ef71bb531b062380f9fd4a9009172fced
+size 37765936
diff --git a/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-100/lora/README.md b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-100/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-100/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-100/lora/adapter_config.json b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-100/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..ec22fc2bfb7e31cdb0828e6cf23b83bbee872655
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-100/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "qkv_proj",
+    "o_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-100/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-100/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..f0ecacc8321adf62e2753a6508837d303339d9bc
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-100/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:40504913201ca48d4b5d89b35ad469f249b0cfbf8665760c6d36cb0f112e322d
+size 37765936
diff --git a/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-110/lora/README.md b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-110/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-110/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-110/lora/adapter_config.json b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-110/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..ec22fc2bfb7e31cdb0828e6cf23b83bbee872655
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-110/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "qkv_proj",
+    "o_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-110/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-110/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..7d536b38d4cffced4e35a02537523cf200a4d754
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-110/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4a9c70e1f17730da5d57f46d7bfb7298e8bd4ebb9aa8010d734152ed2d9df9c6
+size 37765936
diff --git a/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-120/lora/README.md b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-120/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-120/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-120/lora/adapter_config.json b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-120/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..ec22fc2bfb7e31cdb0828e6cf23b83bbee872655
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-120/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "qkv_proj",
+    "o_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-120/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-120/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..5f51ba96b15d313a13808ed3fb2ce06c02e94da2
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-120/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:527a2de86af28c953315c6774156c4b1ccb9be6b2d18a8d8b1e7ead8ba073e3f
+size 37765936
diff --git a/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-130/lora/README.md b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-130/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-130/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-130/lora/adapter_config.json b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-130/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..ec22fc2bfb7e31cdb0828e6cf23b83bbee872655
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-130/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "qkv_proj",
+    "o_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-130/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-130/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..a85e7b171840479563cb646a8c6d9566b6dfc82a
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-130/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1cf2c8ba7a5dd8f91bb24f19935b9c74d59835988e7b16ab930c6673120dce21
+size 37765936
diff --git a/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-140/lora/README.md b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-140/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-140/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-140/lora/adapter_config.json b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-140/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..ec22fc2bfb7e31cdb0828e6cf23b83bbee872655
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-140/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "qkv_proj",
+    "o_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-140/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-140/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..73462b699fbd87dc662d5c0ad4fd0ac55370e4da
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-140/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fcf5959ffe1928d75d950be5c53e5fceae5bbdf5c7436313603d51b565c9d95b
+size 37765936
diff --git a/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-150/lora/README.md b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-150/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-150/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-150/lora/adapter_config.json b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-150/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..ec22fc2bfb7e31cdb0828e6cf23b83bbee872655
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-150/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "qkv_proj",
+    "o_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-150/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-150/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..55a89bb5a23b372b4c563714533244a51697b77f
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-150/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1122412d804b0b6a0a167db2d112fca0245a57cb7bcabfbe1b7753af8bdd13da
+size 37765936
diff --git a/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-160/lora/README.md b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-160/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-160/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-160/lora/adapter_config.json b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-160/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..ec22fc2bfb7e31cdb0828e6cf23b83bbee872655
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-160/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "qkv_proj",
+    "o_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-160/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-160/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..c68fb4f993aa8076c55e5671d0e62550d046de9c
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-160/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:32431ce78dcdc8a85c652d0b9abaeea0269f105718c9f9600a6e973d1e2c094b
+size 37765936
diff --git a/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-170/lora/README.md b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-170/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-170/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-170/lora/adapter_config.json b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-170/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..ec22fc2bfb7e31cdb0828e6cf23b83bbee872655
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-170/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "qkv_proj",
+    "o_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-170/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-170/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..928c06da1043a5f9c3c9512c15e41e9e278d7287
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-170/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1141d6d82042c40892b1389b1aa74fd894926155c47bf556ac0aad43f6a8b44f
+size 37765936
diff --git a/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-180/lora/README.md b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-180/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-180/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-180/lora/adapter_config.json b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-180/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..ec22fc2bfb7e31cdb0828e6cf23b83bbee872655
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-180/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "qkv_proj",
+    "o_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-180/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-180/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..781b7f3657ef668fb397321cb0db14beea43eabf
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-180/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1c67f75e3409993012c8a4e9fbe84ff45085aa7e6393a76a3cead54d9fb44d50
+size 37765936
diff --git a/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-190/lora/README.md b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-190/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-190/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-190/lora/adapter_config.json b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-190/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..ec22fc2bfb7e31cdb0828e6cf23b83bbee872655
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-190/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "qkv_proj",
+    "o_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-190/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-190/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..c69b1c2d0e76514cf16c685c3dd5dd1a4a640603
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-190/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:229261df658fb519c98902d4a66307e2d342dbabe7442e0aa602e5f7fe94b950
+size 37765936
diff --git a/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-20/lora/README.md b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-20/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-20/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-20/lora/adapter_config.json b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-20/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..ec22fc2bfb7e31cdb0828e6cf23b83bbee872655
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-20/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "qkv_proj",
+    "o_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-20/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-20/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..fa44524343993fdad1aef6576d020c98d06eda4b
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-20/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6e45178b280f2ea75547a7708a932610b775caeb55e507f0c8512c0d410347eb
+size 37765936
diff --git a/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-30/lora/README.md b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-30/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-30/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-30/lora/adapter_config.json b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-30/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..ec22fc2bfb7e31cdb0828e6cf23b83bbee872655
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-30/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "qkv_proj",
+    "o_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-30/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-30/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..97314193cb5392f65333e810a492b683b54381af
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-30/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1eb6b9c06fd0e401bf9292403443a873c142e8506abc16d3e5bf6fb92b4cd639
+size 37765936
diff --git a/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-40/lora/README.md b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-40/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-40/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-40/lora/adapter_config.json b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-40/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..ec22fc2bfb7e31cdb0828e6cf23b83bbee872655
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-40/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "qkv_proj",
+    "o_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-40/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-40/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..c8a9a9346600b79a6b2cbfc275ef5f176e43f66e
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-40/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3b9c5992b55a534b10a703d46222289d50dcec9c08489b6256758e437e1698a6
+size 37765936
diff --git a/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-50/lora/README.md b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-50/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-50/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-50/lora/adapter_config.json b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-50/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..ec22fc2bfb7e31cdb0828e6cf23b83bbee872655
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-50/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "qkv_proj",
+    "o_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-50/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-50/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..b3fd514ebad0ab58877fa8bffb5503c1da31983a
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-50/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c81a3919cf71b066d7cfd19c97ddec4d474aaf81aefaa0fbfbf0151910680c96
+size 37765936
diff --git a/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-60/lora/README.md b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-60/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-60/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-60/lora/adapter_config.json b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-60/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..ec22fc2bfb7e31cdb0828e6cf23b83bbee872655
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-60/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "qkv_proj",
+    "o_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-60/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-60/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..e895bb83c0db12bdfdc1713a068c186b6fbb9cc6
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-60/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:331c8cc27521cc5976cf75e7f88f8844b0fcac867cccd5690b5ae9d35aff1075
+size 37765936
diff --git a/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-70/lora/README.md b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-70/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-70/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-70/lora/adapter_config.json b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-70/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..ec22fc2bfb7e31cdb0828e6cf23b83bbee872655
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-70/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "qkv_proj",
+    "o_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-70/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-70/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..0a70be3c522c30450297c19f168b2bee0d5b6b07
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-70/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e1a29f6d16e0fdb481e8ffe9d2b955fbcd0b94dc836bdda0482bbc8674e6934a
+size 37765936
diff --git a/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-80/lora/README.md b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-80/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-80/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-80/lora/adapter_config.json b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-80/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..ec22fc2bfb7e31cdb0828e6cf23b83bbee872655
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-80/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "qkv_proj",
+    "o_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-80/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-80/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..4d17a201b6bb97c599d55ed8b3a3787e2c0b0067
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-80/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:08c72838c16dbca3643cbc73e4b1469b1b03db2348c5dcf7fea08d0d6ebe13e6
+size 37765936
diff --git a/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-90/lora/README.md b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-90/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-90/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-90/lora/adapter_config.json b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-90/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..ec22fc2bfb7e31cdb0828e6cf23b83bbee872655
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-90/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "qkv_proj",
+    "o_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-90/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-90/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..ea252b9853db01d1695c3c2ead5769e010e721d2
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/checkpoints/checkpoint-90/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d749e8634b9fdf4297ecdbf2441ca1ca5caa45406493a9428c2471bd927fa722
+size 37765936
diff --git a/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/environment_summary.txt b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/environment_summary.txt
new file mode 100644
index 0000000000000000000000000000000000000000..566f98f82c027a6dc38095f60310a661cb3afe07
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/environment_summary.txt
@@ -0,0 +1,21 @@
+{
+  "environment": {
+    "CUDA_VISIBLE_DEVICES": "0,1,2,3",
+    "HF_HOME": "/home/wenting/.cache/huggingface",
+    "HF_HUB_CACHE": "/home/wenting/.cache/huggingface/hub",
+    "OMNIGEN_CODE_ROOT": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint",
+    "OUTPUT_DIR": "outputs/fix_win/statcm_win6_10_n001_bs64_steps200",
+    "REWARD_MASK_CHANNELS": "0,1",
+    "SAVE_DIR": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/flow_grpo_cxr/logs/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200",
+    "SFT_LORA_PATH": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint/results_new/scratch_15k",
+    "TEST_JSONL": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_test.jsonl",
+    "TRAIN_JSONL": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_train.jsonl",
+    "WANDB_MODE": "online",
+    "WANDB_NAME": "statcm_win6_10_n001_bs64_h20",
+    "WANDB_PROJECT": "fixwin_h20"
+  },
+  "platform": "Linux-5.4.241-1-tlinux4-0017.7-x86_64-with-glibc2.38",
+  "python": "3.10.20 (main, Mar 11 2026, 17:46:40) [GCC 14.3.0]",
+  "repo_root": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/flow_grpo_cxr",
+  "timestamp": "2026-06-08T13:25:06"
+}
diff --git a/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/git_head.txt b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/git_head.txt
new file mode 100644
index 0000000000000000000000000000000000000000..910c3da6325e61af26279bd74233576c2a0530b2
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/git_head.txt
@@ -0,0 +1 @@
+5c0a81d87d64ffd6ca4e4428794155a290835637
diff --git a/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/git_status.txt b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/git_status.txt
new file mode 100644
index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391
diff --git a/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/launch_config.txt b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/launch_config.txt
new file mode 100644
index 0000000000000000000000000000000000000000..def6d0da18a60af89f17d33d1e419f6ad338410a
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/launch_config.txt
@@ -0,0 +1,49 @@
+timestamp=20260608_132214
+repo_path=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/flow_grpo_cxr
+config_entry=config/grpo.py:general_radiomics_omnigen_4gpu_kl
+server_profile=
+omnigen_code_root=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint
+sft_lora_path=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint/results_new/scratch_15k
+dataset_root=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset
+train_jsonl=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_train.jsonl
+test_jsonl=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_test.jsonl
+dataset_path_remap_from=/home/wenting/dataset/LIDC_IDRI
+dataset_path_remap_to=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset
+mirp_repo_root=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/mirp
+save_dir=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/flow_grpo_cxr/logs/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200
+output_dir=outputs/fix_win/statcm_win6_10_n001_bs64_steps200
+num_steps=10
+eval_num_steps=50
+noise_level=0.01
+sde_type=cps
+sde_window_enabled=1
+sde_window_start=6
+sde_window_end=10
+sde_window_mode=fixed
+sde_window_count=4
+sde_window_train_only=1
+sde_window_logprob_only=0
+SDE_WINDOW_ENABLED=1
+SDE_WINDOW_START=6
+SDE_WINDOW_END=10
+SDE_WINDOW_MODE=fixed
+SDE_WINDOW_COUNT=4
+guidance_scale=2.5
+img_guidance_scale=2.0
+num_image_per_prompt=4
+train_batch_size=64
+test_batch_size=32
+num_batches_per_epoch=2
+kl_beta=0.005
+learning_rate=1e-4
+max_train_steps=200
+reward_feature_prefixes=stat_,cm_
+reward_base_feature_families=stat,cm
+reward_mask_channels=0,1
+num_processes=4
+wandb_project=fixwin_h20
+wandb_group=manual_img_only
+wandb_name=statcm_win6_10_n001_bs64_h20
+wandb_tags=img-only,scratch30k,manual,noise0.01,bs64
+wandb_mode=online
+wandb_resume=never
diff --git a/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/resolved_config.json b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/resolved_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..4d37f772ed221b6f892266b1d9b9eb0a618a86ee
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200/resolved_config.json
@@ -0,0 +1,159 @@
+{
+  "activation_checkpointing": true,
+  "allow_tf32": true,
+  "config_entry": "config/grpo.py:general_radiomics_omnigen_4gpu_kl",
+  "dataset": {
+    "dataset_path_remap_from": "/home/wenting/dataset/LIDC_IDRI",
+    "dataset_path_remap_to": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset",
+    "path_replacements": [
+      [
+        "/home/wenting/dataset/LIDC_IDRI",
+        "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset"
+      ]
+    ],
+    "root": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset",
+    "test_jsonl": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_test.jsonl",
+    "train_jsonl": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_train.jsonl"
+  },
+  "debug": false,
+  "eval_freq": 30,
+  "fsdp_optimizer_offload": false,
+  "logdir": "logs",
+  "mixed_precision": "bf16",
+  "num_checkpoint_limit": 5,
+  "num_epochs": 100000,
+  "per_prompt_stat_tracking": true,
+  "pretrained": {
+    "local_code_root": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint",
+    "model": "Shitao/OmniGen-v1",
+    "revision": "main",
+    "vae_path": null
+  },
+  "prompt_fn": "radiomics_edit",
+  "prompt_fn_kwargs": {},
+  "resolution": 256,
+  "resume_from": null,
+  "reward_fn": {
+    "radiomics": {
+      "base_feature_families": [
+        "stat",
+        "cm"
+      ],
+      "data_root": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset",
+      "dataset_path_remap_from": "/home/wenting/dataset/LIDC_IDRI",
+      "dataset_path_remap_to": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset",
+      "distance_metric": "relative_l1",
+      "feature_prefixes": [
+        "stat_",
+        "cm_"
+      ],
+      "fixed_bin_number": 32,
+      "gt_root": null,
+      "mask_channel": 0,
+      "mask_channels": [
+        0,
+        1
+      ],
+      "mask_root": null,
+      "mirp_repo_root": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/mirp",
+      "missing_score": -10.0,
+      "pixel_spacing_mm": 1.0,
+      "relative_error_clip": 10.0,
+      "reward_scale": 1.0,
+      "roi_name": "lung_roi",
+      "weight": 1.0
+    }
+  },
+  "run_name": "2026.06.08_13.25.02",
+  "sample": {
+    "batch_size": 8,
+    "eval_guidance_scale": 2.5,
+    "eval_img_guidance_scale": 2.0,
+    "eval_num_steps": 50,
+    "global_std": true,
+    "guidance_scale": 2.5,
+    "img_guidance_scale": 2.0,
+    "max_eval_batches": 2,
+    "max_input_image_size": 256,
+    "noise_level": 0.01,
+    "num_batches_per_epoch": 2,
+    "num_image_per_prompt": 4,
+    "num_steps": 10,
+    "same_latent": false,
+    "sde_type": "cps",
+    "sde_window_enabled": true,
+    "sde_window_end": 10,
+    "sde_window_logprob_only": false,
+    "sde_window_mode": "fixed",
+    "sde_window_range": [
+      0,
+      10
+    ],
+    "sde_window_size": 2,
+    "sde_window_start": 6,
+    "sde_window_train_only": true,
+    "test_batch_size": 32,
+    "train_batch_size": 64,
+    "use_img_guidance": true,
+    "use_input_image_size_as_output": false
+  },
+  "save_dir": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/flow_grpo_cxr/logs/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200",
+  "save_freq": 5,
+  "seed": 42,
+  "stage2_extra": {
+    "base_model": "Shitao/OmniGen-v1",
+    "mask_modules_path": null,
+    "omnigen_code_root": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint",
+    "reward_mask_channels": "0,1",
+    "rl_lora_target_modules": [
+      "qkv_proj",
+      "o_proj"
+    ],
+    "save_dir": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/flow_grpo_cxr/logs/radiomics/fix_win/statcm_win6_10_n001_bs64_steps200",
+    "sde_window_enabled": true,
+    "sde_window_end": 10,
+    "sde_window_mode": "fixed",
+    "sde_window_start": 6,
+    "sft_lora_path": null,
+    "test_jsonl": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_test.jsonl",
+    "train_jsonl": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_train.jsonl",
+    "train_timestep_indices": [
+      6,
+      7,
+      8,
+      9
+    ]
+  },
+  "train": {
+    "adam_beta1": 0.9,
+    "adam_beta2": 0.999,
+    "adam_epsilon": 1e-08,
+    "adam_weight_decay": 0.0001,
+    "adv_clip_max": 5,
+    "batch_size": 64,
+    "beta": 0.005,
+    "cfg": true,
+    "clip_range": 0.0001,
+    "clip_range_gt": 0.0001,
+    "clip_range_lt": 0.0001,
+    "condition_dropout_prob": 0.01,
+    "ema": false,
+    "gradient_accumulation_steps": 1,
+    "learning_rate": 0.0001,
+    "lora_alpha": 64,
+    "lora_path": null,
+    "lora_rank": 32,
+    "max_grad_norm": 1.0,
+    "max_train_steps": 200,
+    "merge_lora_path": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint/results_new/scratch_15k",
+    "num_inner_epochs": 1,
+    "rl_lora_target_modules": [
+      "qkv_proj",
+      "o_proj"
+    ],
+    "timestep_fraction": 1.0,
+    "timestep_shift": 3.0,
+    "use_8bit_adam": false
+  },
+  "use_lora": true
+}
diff --git a/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-10/lora/README.md b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-10/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-10/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-10/lora/adapter_config.json b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-10/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..5caca2d3fe4f842c2d173e65caa2fb5f52a00540
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-10/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "qkv_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-10/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-10/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..8ff4c4c3c486e0ef39a6fe95d31a97f8682bb156
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-10/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:43da90852b32d61c5842f5bd2cba26772e86147dce5a9a909f3231a4cb2e5711
+size 37765936
diff --git a/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-100/lora/README.md b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-100/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-100/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-100/lora/adapter_config.json b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-100/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..5caca2d3fe4f842c2d173e65caa2fb5f52a00540
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-100/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "qkv_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-100/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-100/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..857909c26217bb85fbcb8ddc97bbfb871e25efbe
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-100/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:41be5c393cb26465b571a291c97e9b5ab26a0f2e9c052124318b3ec9372f700b
+size 37765936
diff --git a/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-110/lora/README.md b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-110/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-110/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-110/lora/adapter_config.json b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-110/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..5caca2d3fe4f842c2d173e65caa2fb5f52a00540
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-110/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "qkv_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-110/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-110/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..1ddbb27cf4faaafe3460326f0b84531cdd75c3db
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-110/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5667426de38d529f6dd7673b5c280949370c588540f9b028e9f8dcac57c345fc
+size 37765936
diff --git a/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-120/lora/README.md b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-120/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-120/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-120/lora/adapter_config.json b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-120/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..5caca2d3fe4f842c2d173e65caa2fb5f52a00540
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-120/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "qkv_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-120/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-120/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..195230b10c4ddff0dbe08e52d35328559f5c32c7
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-120/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:28c6b2e670155fae17d98e689f04647f508777e1afb2bfa70b34fb636df7b7bc
+size 37765936
diff --git a/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-130/lora/README.md b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-130/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-130/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-130/lora/adapter_config.json b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-130/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..5caca2d3fe4f842c2d173e65caa2fb5f52a00540
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-130/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "qkv_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-130/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-130/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..9fcff21d1d1b713e434c662c657fec61bc6a150e
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-130/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61bea8ea040bc3e4a00b6bbb2ae79ee60bc781b1d6400cfb523d11c191e2ca86
+size 37765936
diff --git a/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-140/lora/README.md b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-140/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-140/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-140/lora/adapter_config.json b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-140/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..5caca2d3fe4f842c2d173e65caa2fb5f52a00540
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-140/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "qkv_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-140/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-140/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..eef5cdd7ea977badf02b24e46fed5cb5642ce01a
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-140/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:33a30650c88f4c31d252c093163abec4279679141bb51241acfad7c9130d336b
+size 37765936
diff --git a/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-150/lora/README.md b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-150/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-150/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-150/lora/adapter_config.json b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-150/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..5caca2d3fe4f842c2d173e65caa2fb5f52a00540
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-150/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "qkv_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-150/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-150/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..a1726b539306c6f5c34bdac189b03571a8a5a671
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-150/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ced134570f39d5454953fb204855e6bf38823a5f6e996eae87fb8e3f27ffed7a
+size 37765936
diff --git a/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-160/lora/README.md b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-160/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-160/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-160/lora/adapter_config.json b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-160/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..5caca2d3fe4f842c2d173e65caa2fb5f52a00540
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-160/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "qkv_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-160/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-160/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..0021201f0c2b58e702a55e1dc2c28044b631a197
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-160/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7a7f6eb9476d02523a9f7664c3b214901737f07e94fdb1ed9404225098c6a6e3
+size 37765936
diff --git a/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-170/lora/README.md b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-170/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-170/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-170/lora/adapter_config.json b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-170/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..5caca2d3fe4f842c2d173e65caa2fb5f52a00540
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-170/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "qkv_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-170/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-170/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..d66542ee1d2e206f5c65d47b38c95dd633452bc3
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-170/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:885a6205c73bf2151e1b43363444dc3bd7795954e92926c378bb0c447396e7ca
+size 37765936
diff --git a/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-180/lora/README.md b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-180/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-180/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-180/lora/adapter_config.json b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-180/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..5caca2d3fe4f842c2d173e65caa2fb5f52a00540
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-180/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "qkv_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-180/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-180/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..589b5d3cb15dc315544388a9b5e05e68f1b820d9
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-180/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:39c2c1ba3c06d9eb27ca8bbfde1539930dd1e2c15f7fc2c365467ae1975651e2
+size 37765936
diff --git a/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-190/lora/README.md b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-190/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-190/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-190/lora/adapter_config.json b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-190/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..5caca2d3fe4f842c2d173e65caa2fb5f52a00540
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-190/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "qkv_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-190/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-190/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..df2bc9137a795793a475d9964d28d955ede05eb4
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-190/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:236174201b9d24d7efc7b1d416bbf4e27fc0c8618e113600fef5964d25ae8452
+size 37765936
diff --git a/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-20/lora/README.md b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-20/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-20/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-20/lora/adapter_config.json b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-20/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..5caca2d3fe4f842c2d173e65caa2fb5f52a00540
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-20/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "qkv_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-20/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-20/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..754fa23b68671ccf0fb3ec93291969ffa8a0624c
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-20/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d9dcf92d69c9402852f438f73f1f2e6e11719d8f4fae9d6fae6293cd91e9cd62
+size 37765936
diff --git a/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-30/lora/README.md b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-30/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-30/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-30/lora/adapter_config.json b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-30/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..5caca2d3fe4f842c2d173e65caa2fb5f52a00540
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-30/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "qkv_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-30/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-30/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..f9cfaa0d343e6b9d8682c561cc1d92348e839969
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-30/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f993b3f7f96752f01f3e1ca458d11d795ffd4d5e5151ee2c83ec164f57314ef5
+size 37765936
diff --git a/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-40/lora/README.md b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-40/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-40/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-40/lora/adapter_config.json b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-40/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..5caca2d3fe4f842c2d173e65caa2fb5f52a00540
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-40/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "qkv_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-40/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-40/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..0d141edbdc290413c44de369e0c546aa40e19167
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-40/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d8a92a4180f6affa7bc28fc8016845921986e9f2d69730e64b92bbc0541dba01
+size 37765936
diff --git a/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-50/lora/README.md b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-50/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-50/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-50/lora/adapter_config.json b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-50/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..5caca2d3fe4f842c2d173e65caa2fb5f52a00540
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-50/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "qkv_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-50/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-50/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..e223199b5330157724031708fb401f9ec2bbaa3a
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-50/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bfaab0b2852cf8d98728c32d61bc79d80b38636ffab762d56b6665da8ed8c6bf
+size 37765936
diff --git a/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-60/lora/README.md b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-60/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-60/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-60/lora/adapter_config.json b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-60/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..5caca2d3fe4f842c2d173e65caa2fb5f52a00540
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-60/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "qkv_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-60/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-60/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..b9a22f124443ac3c8dbc41d24ef3ba7890208186
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-60/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:323c92cf248fa5794ae6c10d2d94bf475fdaac882a59dd4dca727f9fe54dca10
+size 37765936
diff --git a/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-70/lora/README.md b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-70/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-70/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-70/lora/adapter_config.json b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-70/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..5caca2d3fe4f842c2d173e65caa2fb5f52a00540
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-70/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "qkv_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-70/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-70/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..27b346ae525ddab6dff14f84958f42002d4386a7
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-70/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:028ce39f02366236af2ca1b4cddde8434f67661483385ef145cf0859afcb3ede
+size 37765936
diff --git a/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-80/lora/README.md b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-80/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-80/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-80/lora/adapter_config.json b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-80/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..5caca2d3fe4f842c2d173e65caa2fb5f52a00540
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-80/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "qkv_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-80/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-80/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..dd6e48cdec28b3d23777beea9225f1edc5f6837c
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-80/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:86b2298481e9a654d770623734778b52fe0814a33bdcbc14d87426fe81198c57
+size 37765936
diff --git a/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-90/lora/README.md b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-90/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-90/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-90/lora/adapter_config.json b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-90/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..5caca2d3fe4f842c2d173e65caa2fb5f52a00540
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-90/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "qkv_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-90/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-90/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..e4a0d536193a55cdecefe155d28893fca376d595
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/checkpoints/checkpoint-90/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9c1c34ae18d74d8f60b859072dc913dcbbdce11f4128080ae1dd63951bf224f3
+size 37765936
diff --git a/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/environment_summary.txt b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/environment_summary.txt
new file mode 100644
index 0000000000000000000000000000000000000000..3cdab3a3e7df3fe75b9071bed8e39e7952de4180
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/environment_summary.txt
@@ -0,0 +1,21 @@
+{
+  "environment": {
+    "CUDA_VISIBLE_DEVICES": "4,5,6,7",
+    "HF_HOME": "/home/wenting/.cache/huggingface",
+    "HF_HUB_CACHE": "/home/wenting/.cache/huggingface/hub",
+    "OMNIGEN_CODE_ROOT": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint",
+    "OUTPUT_DIR": "outputs/fix_win/statcm_win6_8_n001_bs64_steps200",
+    "REWARD_MASK_CHANNELS": "0,1",
+    "SAVE_DIR": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/flow_grpo_cxr/logs/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200",
+    "SFT_LORA_PATH": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint/results_new/scratch_15k",
+    "TEST_JSONL": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_test.jsonl",
+    "TRAIN_JSONL": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_train.jsonl",
+    "WANDB_MODE": "online",
+    "WANDB_NAME": "statcm_win6_8_n001_bs64_h20",
+    "WANDB_PROJECT": "fixwin_h20"
+  },
+  "platform": "Linux-5.4.241-1-tlinux4-0017.7-x86_64-with-glibc2.38",
+  "python": "3.10.20 (main, Mar 11 2026, 17:46:40) [GCC 14.3.0]",
+  "repo_root": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/flow_grpo_cxr",
+  "timestamp": "2026-06-08T13:31:00"
+}
diff --git a/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/git_head.txt b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/git_head.txt
new file mode 100644
index 0000000000000000000000000000000000000000..910c3da6325e61af26279bd74233576c2a0530b2
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/git_head.txt
@@ -0,0 +1 @@
+5c0a81d87d64ffd6ca4e4428794155a290835637
diff --git a/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/git_status.txt b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/git_status.txt
new file mode 100644
index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391
diff --git a/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/launch_config.txt b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/launch_config.txt
new file mode 100644
index 0000000000000000000000000000000000000000..5dd0b888133ab612b92fcb6795064c74c7e24330
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/launch_config.txt
@@ -0,0 +1,49 @@
+timestamp=20260608_132801
+repo_path=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/flow_grpo_cxr
+config_entry=config/grpo.py:general_radiomics_omnigen_4gpu_kl
+server_profile=
+omnigen_code_root=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint
+sft_lora_path=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint/results_new/scratch_15k
+dataset_root=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset
+train_jsonl=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_train.jsonl
+test_jsonl=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_test.jsonl
+dataset_path_remap_from=/home/wenting/dataset/LIDC_IDRI
+dataset_path_remap_to=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset
+mirp_repo_root=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/mirp
+save_dir=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/flow_grpo_cxr/logs/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200
+output_dir=outputs/fix_win/statcm_win6_8_n001_bs64_steps200
+num_steps=10
+eval_num_steps=50
+noise_level=0.01
+sde_type=cps
+sde_window_enabled=1
+sde_window_start=6
+sde_window_end=8
+sde_window_mode=fixed
+sde_window_count=2
+sde_window_train_only=1
+sde_window_logprob_only=0
+SDE_WINDOW_ENABLED=1
+SDE_WINDOW_START=6
+SDE_WINDOW_END=8
+SDE_WINDOW_MODE=fixed
+SDE_WINDOW_COUNT=2
+guidance_scale=2.5
+img_guidance_scale=2.0
+num_image_per_prompt=4
+train_batch_size=64
+test_batch_size=32
+num_batches_per_epoch=2
+kl_beta=0.005
+learning_rate=1e-4
+max_train_steps=200
+reward_feature_prefixes=stat_,cm_
+reward_base_feature_families=stat,cm
+reward_mask_channels=0,1
+num_processes=4
+wandb_project=fixwin_h20
+wandb_group=manual_img_only
+wandb_name=statcm_win6_8_n001_bs64_h20
+wandb_tags=img-only,scratch30k,manual,noise0.01,bs64
+wandb_mode=online
+wandb_resume=never
diff --git a/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/resolved_config.json b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/resolved_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..9702f404a9b8d4b94b75c2d360381cf633389e26
--- /dev/null
+++ b/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200/resolved_config.json
@@ -0,0 +1,157 @@
+{
+  "activation_checkpointing": true,
+  "allow_tf32": true,
+  "config_entry": "config/grpo.py:general_radiomics_omnigen_4gpu_kl",
+  "dataset": {
+    "dataset_path_remap_from": "/home/wenting/dataset/LIDC_IDRI",
+    "dataset_path_remap_to": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset",
+    "path_replacements": [
+      [
+        "/home/wenting/dataset/LIDC_IDRI",
+        "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset"
+      ]
+    ],
+    "root": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset",
+    "test_jsonl": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_test.jsonl",
+    "train_jsonl": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_train.jsonl"
+  },
+  "debug": false,
+  "eval_freq": 30,
+  "fsdp_optimizer_offload": false,
+  "logdir": "logs",
+  "mixed_precision": "bf16",
+  "num_checkpoint_limit": 5,
+  "num_epochs": 100000,
+  "per_prompt_stat_tracking": true,
+  "pretrained": {
+    "local_code_root": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint",
+    "model": "Shitao/OmniGen-v1",
+    "revision": "main",
+    "vae_path": null
+  },
+  "prompt_fn": "radiomics_edit",
+  "prompt_fn_kwargs": {},
+  "resolution": 256,
+  "resume_from": null,
+  "reward_fn": {
+    "radiomics": {
+      "base_feature_families": [
+        "stat",
+        "cm"
+      ],
+      "data_root": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset",
+      "dataset_path_remap_from": "/home/wenting/dataset/LIDC_IDRI",
+      "dataset_path_remap_to": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset",
+      "distance_metric": "relative_l1",
+      "feature_prefixes": [
+        "stat_",
+        "cm_"
+      ],
+      "fixed_bin_number": 32,
+      "gt_root": null,
+      "mask_channel": 0,
+      "mask_channels": [
+        0,
+        1
+      ],
+      "mask_root": null,
+      "mirp_repo_root": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/mirp",
+      "missing_score": -10.0,
+      "pixel_spacing_mm": 1.0,
+      "relative_error_clip": 10.0,
+      "reward_scale": 1.0,
+      "roi_name": "lung_roi",
+      "weight": 1.0
+    }
+  },
+  "run_name": "2026.06.08_13.30.56",
+  "sample": {
+    "batch_size": 8,
+    "eval_guidance_scale": 2.5,
+    "eval_img_guidance_scale": 2.0,
+    "eval_num_steps": 50,
+    "global_std": true,
+    "guidance_scale": 2.5,
+    "img_guidance_scale": 2.0,
+    "max_eval_batches": 2,
+    "max_input_image_size": 256,
+    "noise_level": 0.01,
+    "num_batches_per_epoch": 2,
+    "num_image_per_prompt": 4,
+    "num_steps": 10,
+    "same_latent": false,
+    "sde_type": "cps",
+    "sde_window_enabled": true,
+    "sde_window_end": 8,
+    "sde_window_logprob_only": false,
+    "sde_window_mode": "fixed",
+    "sde_window_range": [
+      0,
+      10
+    ],
+    "sde_window_size": 2,
+    "sde_window_start": 6,
+    "sde_window_train_only": true,
+    "test_batch_size": 32,
+    "train_batch_size": 64,
+    "use_img_guidance": true,
+    "use_input_image_size_as_output": false
+  },
+  "save_dir": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/flow_grpo_cxr/logs/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200",
+  "save_freq": 5,
+  "seed": 42,
+  "stage2_extra": {
+    "base_model": "Shitao/OmniGen-v1",
+    "mask_modules_path": null,
+    "omnigen_code_root": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint",
+    "reward_mask_channels": "0,1",
+    "rl_lora_target_modules": [
+      "qkv_proj",
+      "o_proj"
+    ],
+    "save_dir": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/flow_grpo_cxr/logs/radiomics/fix_win/statcm_win6_8_n001_bs64_steps200",
+    "sde_window_enabled": true,
+    "sde_window_end": 8,
+    "sde_window_mode": "fixed",
+    "sde_window_start": 6,
+    "sft_lora_path": null,
+    "test_jsonl": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_test.jsonl",
+    "train_jsonl": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_train.jsonl",
+    "train_timestep_indices": [
+      6,
+      7
+    ]
+  },
+  "train": {
+    "adam_beta1": 0.9,
+    "adam_beta2": 0.999,
+    "adam_epsilon": 1e-08,
+    "adam_weight_decay": 0.0001,
+    "adv_clip_max": 5,
+    "batch_size": 64,
+    "beta": 0.005,
+    "cfg": true,
+    "clip_range": 0.0001,
+    "clip_range_gt": 0.0001,
+    "clip_range_lt": 0.0001,
+    "condition_dropout_prob": 0.01,
+    "ema": false,
+    "gradient_accumulation_steps": 1,
+    "learning_rate": 0.0001,
+    "lora_alpha": 64,
+    "lora_path": null,
+    "lora_rank": 32,
+    "max_grad_norm": 1.0,
+    "max_train_steps": 200,
+    "merge_lora_path": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint/results_new/scratch_15k",
+    "num_inner_epochs": 1,
+    "rl_lora_target_modules": [
+      "qkv_proj",
+      "o_proj"
+    ],
+    "timestep_fraction": 1.0,
+    "timestep_shift": 3.0,
+    "use_8bit_adam": false
+  },
+  "use_lora": true
+}
diff --git a/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-10/lora/README.md b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-10/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-10/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-10/lora/adapter_config.json b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-10/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..5caca2d3fe4f842c2d173e65caa2fb5f52a00540
--- /dev/null
+++ b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-10/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "qkv_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-10/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-10/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..9d5d04ae723622b01c35bc14b394be32d6df3e29
--- /dev/null
+++ b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-10/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cc81719a64568f7ea545072cb4699a37c3a83ac8a4d0aebedcc4b33ce1366440
+size 37765936
diff --git a/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-100/lora/README.md b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-100/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-100/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-100/lora/adapter_config.json b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-100/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..5caca2d3fe4f842c2d173e65caa2fb5f52a00540
--- /dev/null
+++ b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-100/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "qkv_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-100/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-100/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..02886475160538469c1ddd98bab1439fab61c390
--- /dev/null
+++ b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-100/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:da8484ebcbb6a057e5b451f5484e7b3fdf2babf58bf4c5024aa5dae8742d5105
+size 37765936
diff --git a/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-110/lora/README.md b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-110/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-110/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-110/lora/adapter_config.json b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-110/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..5caca2d3fe4f842c2d173e65caa2fb5f52a00540
--- /dev/null
+++ b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-110/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "qkv_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-110/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-110/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..f000c5af8c7129304c985c6acb3c7b2a8c9305e3
--- /dev/null
+++ b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-110/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:888e6e2eb8ba0f00bc0adccd6e90b0e894b6173560e066ef7ac33b7165b73176
+size 37765936
diff --git a/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-120/lora/README.md b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-120/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-120/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-120/lora/adapter_config.json b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-120/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..5caca2d3fe4f842c2d173e65caa2fb5f52a00540
--- /dev/null
+++ b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-120/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "qkv_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-120/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-120/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..101bc3514122ed269f85985ee75c1e400f30ffb3
--- /dev/null
+++ b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-120/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5c0d57908dfd2a102742d3bf1b80028b5fadd4f54340709d7ff50fbaed0921f1
+size 37765936
diff --git a/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-130/lora/README.md b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-130/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-130/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-130/lora/adapter_config.json b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-130/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..5caca2d3fe4f842c2d173e65caa2fb5f52a00540
--- /dev/null
+++ b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-130/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "qkv_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-130/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-130/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..1c64d79d1e8f434005f62afb627f1bbe2ad71701
--- /dev/null
+++ b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-130/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a8c2059f9bc0bc17d52606cf9be4e88cb03e5c3be5dabaaa097a8dcaab30c341
+size 37765936
diff --git a/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-140/lora/README.md b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-140/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-140/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-140/lora/adapter_config.json b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-140/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..5caca2d3fe4f842c2d173e65caa2fb5f52a00540
--- /dev/null
+++ b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-140/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "qkv_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-140/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-140/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..a46f6b4f60c66c756ca53bb9215c6241621b0b35
--- /dev/null
+++ b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-140/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6dfa5b08faafaaabed011f8c97505fefecdcb2e19957385a8a318aa9c413e4b3
+size 37765936
diff --git a/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-150/lora/README.md b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-150/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-150/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-150/lora/adapter_config.json b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-150/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..5caca2d3fe4f842c2d173e65caa2fb5f52a00540
--- /dev/null
+++ b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-150/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "qkv_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-150/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-150/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..82636ebed43e4f850f23cc9f5cffddf665da9cec
--- /dev/null
+++ b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-150/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e2011a864f482af0849cd59c79b2468e2520f61af35021c71a6b5ecbc80e766f
+size 37765936
diff --git a/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-160/lora/README.md b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-160/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-160/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-160/lora/adapter_config.json b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-160/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..5caca2d3fe4f842c2d173e65caa2fb5f52a00540
--- /dev/null
+++ b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-160/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "qkv_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-160/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-160/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..d7eb10bb5f424acd5f2106746392c8d8954c9a87
--- /dev/null
+++ b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-160/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c22b8a693c4b96cb95ffddc38a9cb1ead125c0227fbbf0533bba94a1582b5af3
+size 37765936
diff --git a/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-170/lora/README.md b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-170/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-170/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-170/lora/adapter_config.json b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-170/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..5caca2d3fe4f842c2d173e65caa2fb5f52a00540
--- /dev/null
+++ b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-170/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "qkv_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-170/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-170/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..38a39b0538da06e2ec04230ec803c5e8d1ef7c38
--- /dev/null
+++ b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-170/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:34b288fc1a79e105586bf6bf26b470c66b0f6bfae634b0b2cececca3ace89871
+size 37765936
diff --git a/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-180/lora/README.md b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-180/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-180/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-180/lora/adapter_config.json b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-180/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..5caca2d3fe4f842c2d173e65caa2fb5f52a00540
--- /dev/null
+++ b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-180/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "qkv_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-180/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-180/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..9cafe71e3db0007fb4f2392aa161707b8c9341a2
--- /dev/null
+++ b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-180/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e7db93d19087c162ff5648a3fd2814d9ab542874bb9e4aa3d5d8d08e65072bac
+size 37765936
diff --git a/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-190/lora/README.md b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-190/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-190/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-190/lora/adapter_config.json b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-190/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..5caca2d3fe4f842c2d173e65caa2fb5f52a00540
--- /dev/null
+++ b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-190/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "qkv_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-190/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-190/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..56beed8918a38ed69bcc3d2c292fed3a27f7f234
--- /dev/null
+++ b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-190/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d9739b4f36869c0316995e66c7f97990370ee642d3866e092ab95d43411b267b
+size 37765936
diff --git a/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-20/lora/README.md b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-20/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-20/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-20/lora/adapter_config.json b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-20/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..5caca2d3fe4f842c2d173e65caa2fb5f52a00540
--- /dev/null
+++ b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-20/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "qkv_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-20/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-20/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..6e429506fd75e7fc0ce2fb4b10bbf1033aa492b7
--- /dev/null
+++ b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-20/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6b54179d114b1640d3c47b569799ecdfc9b2028bf7d826ee63c17d23d9e0b3a6
+size 37765936
diff --git a/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-30/lora/README.md b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-30/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-30/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-30/lora/adapter_config.json b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-30/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..5caca2d3fe4f842c2d173e65caa2fb5f52a00540
--- /dev/null
+++ b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-30/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "qkv_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-30/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-30/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..2ace0dd25f0f07b2ff15d09f43514eed736c4eab
--- /dev/null
+++ b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-30/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b1a4ce97a127d0077c57fffcc047c59edc3886eb98f363aef99a1ede9f84e707
+size 37765936
diff --git a/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-40/lora/README.md b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-40/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-40/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-40/lora/adapter_config.json b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-40/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..5caca2d3fe4f842c2d173e65caa2fb5f52a00540
--- /dev/null
+++ b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-40/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "qkv_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-40/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-40/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..acde213bafa58774007c55b7c831d00216c691dd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-40/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1251b8276112b1870e80fb3c49813ed528033f7d83cd94392d39a4465915e35a
+size 37765936
diff --git a/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-50/lora/README.md b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-50/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-50/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-50/lora/adapter_config.json b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-50/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..5caca2d3fe4f842c2d173e65caa2fb5f52a00540
--- /dev/null
+++ b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-50/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "qkv_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-50/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-50/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..90da0a50eade005846bd4fb711361000d534b615
--- /dev/null
+++ b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-50/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:60605af64c37a571767a8fb87a3f2f8539e3c8a8afba252986963e5eea20f046
+size 37765936
diff --git a/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-60/lora/README.md b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-60/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-60/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-60/lora/adapter_config.json b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-60/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..5caca2d3fe4f842c2d173e65caa2fb5f52a00540
--- /dev/null
+++ b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-60/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "qkv_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-60/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-60/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..eb039dc5719001d2d8a9c19e430b04bff0856948
--- /dev/null
+++ b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-60/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:10579546fb71437526d3b06887eda4a82054aebc4aca711f707d941206cf46bc
+size 37765936
diff --git a/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-70/lora/README.md b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-70/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-70/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-70/lora/adapter_config.json b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-70/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..5caca2d3fe4f842c2d173e65caa2fb5f52a00540
--- /dev/null
+++ b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-70/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "qkv_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-70/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-70/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..fd397102a9c9487f965792822ab56becdfcfee8f
--- /dev/null
+++ b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-70/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:960737a7da28f91914529bdf1e3cff4c63882edec4059d3cee24a0eb44315fe4
+size 37765936
diff --git a/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-80/lora/README.md b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-80/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-80/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-80/lora/adapter_config.json b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-80/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..5caca2d3fe4f842c2d173e65caa2fb5f52a00540
--- /dev/null
+++ b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-80/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "qkv_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-80/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-80/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..46c820ce194013ef5ddc2d43728a4dcde289027a
--- /dev/null
+++ b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-80/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b2af4ff84bc6273d36f26884d00bd3f2720b3520d8331775fb5e7d8a9198fc59
+size 37765936
diff --git a/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-90/lora/README.md b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-90/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-90/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-90/lora/adapter_config.json b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-90/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..5caca2d3fe4f842c2d173e65caa2fb5f52a00540
--- /dev/null
+++ b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-90/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "qkv_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-90/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-90/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..afd45032268f7d36855846f901212976f914b498
--- /dev/null
+++ b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/checkpoints/checkpoint-90/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b84b6e79f5d051313c05678820821fdd065f7288691fa80bd9384c35705245fb
+size 37765936
diff --git a/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/environment_summary.txt b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/environment_summary.txt
new file mode 100644
index 0000000000000000000000000000000000000000..d31decf75303d106214ab27105e9e21181780f7f
--- /dev/null
+++ b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/environment_summary.txt
@@ -0,0 +1,21 @@
+{
+  "environment": {
+    "CUDA_VISIBLE_DEVICES": "4,5,6,7",
+    "HF_HOME": "/home/wenting/.cache/huggingface",
+    "HF_HUB_CACHE": "/home/wenting/.cache/huggingface/hub",
+    "OMNIGEN_CODE_ROOT": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint",
+    "OUTPUT_DIR": "outputs/fix_win/statcm_win7_10_n001_bs64_steps200",
+    "REWARD_MASK_CHANNELS": "0,1",
+    "SAVE_DIR": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/flow_grpo_cxr/logs/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200",
+    "SFT_LORA_PATH": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint/results_new/scratch_15k",
+    "TEST_JSONL": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_test.jsonl",
+    "TRAIN_JSONL": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_train.jsonl",
+    "WANDB_MODE": "online",
+    "WANDB_NAME": "statcm_win7_10_n001_bs64_h20",
+    "WANDB_PROJECT": "fixwin_h20"
+  },
+  "platform": "Linux-5.4.241-1-tlinux4-0017.7-x86_64-with-glibc2.38",
+  "python": "3.10.20 (main, Mar 11 2026, 17:46:40) [GCC 14.3.0]",
+  "repo_root": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/flow_grpo_cxr",
+  "timestamp": "2026-06-08T13:27:10"
+}
diff --git a/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/git_head.txt b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/git_head.txt
new file mode 100644
index 0000000000000000000000000000000000000000..910c3da6325e61af26279bd74233576c2a0530b2
--- /dev/null
+++ b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/git_head.txt
@@ -0,0 +1 @@
+5c0a81d87d64ffd6ca4e4428794155a290835637
diff --git a/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/git_status.txt b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/git_status.txt
new file mode 100644
index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391
diff --git a/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/launch_config.txt b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/launch_config.txt
new file mode 100644
index 0000000000000000000000000000000000000000..eac2c33b325fc6626bb653b412e6f65a72d4c109
--- /dev/null
+++ b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/launch_config.txt
@@ -0,0 +1,49 @@
+timestamp=20260608_132414
+repo_path=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/flow_grpo_cxr
+config_entry=config/grpo.py:general_radiomics_omnigen_4gpu_kl
+server_profile=
+omnigen_code_root=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint
+sft_lora_path=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint/results_new/scratch_15k
+dataset_root=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset
+train_jsonl=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_train.jsonl
+test_jsonl=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_test.jsonl
+dataset_path_remap_from=/home/wenting/dataset/LIDC_IDRI
+dataset_path_remap_to=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset
+mirp_repo_root=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/mirp
+save_dir=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/flow_grpo_cxr/logs/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200
+output_dir=outputs/fix_win/statcm_win7_10_n001_bs64_steps200
+num_steps=10
+eval_num_steps=50
+noise_level=0.01
+sde_type=cps
+sde_window_enabled=1
+sde_window_start=7
+sde_window_end=10
+sde_window_mode=fixed
+sde_window_count=3
+sde_window_train_only=1
+sde_window_logprob_only=0
+SDE_WINDOW_ENABLED=1
+SDE_WINDOW_START=7
+SDE_WINDOW_END=10
+SDE_WINDOW_MODE=fixed
+SDE_WINDOW_COUNT=3
+guidance_scale=2.5
+img_guidance_scale=2.0
+num_image_per_prompt=4
+train_batch_size=64
+test_batch_size=32
+num_batches_per_epoch=2
+kl_beta=0.005
+learning_rate=1e-4
+max_train_steps=200
+reward_feature_prefixes=stat_,cm_
+reward_base_feature_families=stat,cm
+reward_mask_channels=0,1
+num_processes=4
+wandb_project=fixwin_h20
+wandb_group=manual_img_only
+wandb_name=statcm_win7_10_n001_bs64_h20
+wandb_tags=img-only,scratch30k,manual,noise0.01,bs64
+wandb_mode=online
+wandb_resume=never
diff --git a/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/resolved_config.json b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/resolved_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..bdae9d92f70dc87901b009477fa14b3d23e3957d
--- /dev/null
+++ b/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200/resolved_config.json
@@ -0,0 +1,158 @@
+{
+  "activation_checkpointing": true,
+  "allow_tf32": true,
+  "config_entry": "config/grpo.py:general_radiomics_omnigen_4gpu_kl",
+  "dataset": {
+    "dataset_path_remap_from": "/home/wenting/dataset/LIDC_IDRI",
+    "dataset_path_remap_to": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset",
+    "path_replacements": [
+      [
+        "/home/wenting/dataset/LIDC_IDRI",
+        "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset"
+      ]
+    ],
+    "root": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset",
+    "test_jsonl": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_test.jsonl",
+    "train_jsonl": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_train.jsonl"
+  },
+  "debug": false,
+  "eval_freq": 30,
+  "fsdp_optimizer_offload": false,
+  "logdir": "logs",
+  "mixed_precision": "bf16",
+  "num_checkpoint_limit": 5,
+  "num_epochs": 100000,
+  "per_prompt_stat_tracking": true,
+  "pretrained": {
+    "local_code_root": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint",
+    "model": "Shitao/OmniGen-v1",
+    "revision": "main",
+    "vae_path": null
+  },
+  "prompt_fn": "radiomics_edit",
+  "prompt_fn_kwargs": {},
+  "resolution": 256,
+  "resume_from": null,
+  "reward_fn": {
+    "radiomics": {
+      "base_feature_families": [
+        "stat",
+        "cm"
+      ],
+      "data_root": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset",
+      "dataset_path_remap_from": "/home/wenting/dataset/LIDC_IDRI",
+      "dataset_path_remap_to": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset",
+      "distance_metric": "relative_l1",
+      "feature_prefixes": [
+        "stat_",
+        "cm_"
+      ],
+      "fixed_bin_number": 32,
+      "gt_root": null,
+      "mask_channel": 0,
+      "mask_channels": [
+        0,
+        1
+      ],
+      "mask_root": null,
+      "mirp_repo_root": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/mirp",
+      "missing_score": -10.0,
+      "pixel_spacing_mm": 1.0,
+      "relative_error_clip": 10.0,
+      "reward_scale": 1.0,
+      "roi_name": "lung_roi",
+      "weight": 1.0
+    }
+  },
+  "run_name": "2026.06.08_13.27.07",
+  "sample": {
+    "batch_size": 8,
+    "eval_guidance_scale": 2.5,
+    "eval_img_guidance_scale": 2.0,
+    "eval_num_steps": 50,
+    "global_std": true,
+    "guidance_scale": 2.5,
+    "img_guidance_scale": 2.0,
+    "max_eval_batches": 2,
+    "max_input_image_size": 256,
+    "noise_level": 0.01,
+    "num_batches_per_epoch": 2,
+    "num_image_per_prompt": 4,
+    "num_steps": 10,
+    "same_latent": false,
+    "sde_type": "cps",
+    "sde_window_enabled": true,
+    "sde_window_end": 10,
+    "sde_window_logprob_only": false,
+    "sde_window_mode": "fixed",
+    "sde_window_range": [
+      0,
+      10
+    ],
+    "sde_window_size": 2,
+    "sde_window_start": 7,
+    "sde_window_train_only": true,
+    "test_batch_size": 32,
+    "train_batch_size": 64,
+    "use_img_guidance": true,
+    "use_input_image_size_as_output": false
+  },
+  "save_dir": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/flow_grpo_cxr/logs/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200",
+  "save_freq": 5,
+  "seed": 42,
+  "stage2_extra": {
+    "base_model": "Shitao/OmniGen-v1",
+    "mask_modules_path": null,
+    "omnigen_code_root": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint",
+    "reward_mask_channels": "0,1",
+    "rl_lora_target_modules": [
+      "qkv_proj",
+      "o_proj"
+    ],
+    "save_dir": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/flow_grpo_cxr/logs/radiomics/fix_win/statcm_win7_10_n001_bs64_steps200",
+    "sde_window_enabled": true,
+    "sde_window_end": 10,
+    "sde_window_mode": "fixed",
+    "sde_window_start": 7,
+    "sft_lora_path": null,
+    "test_jsonl": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_test.jsonl",
+    "train_jsonl": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_train.jsonl",
+    "train_timestep_indices": [
+      7,
+      8,
+      9
+    ]
+  },
+  "train": {
+    "adam_beta1": 0.9,
+    "adam_beta2": 0.999,
+    "adam_epsilon": 1e-08,
+    "adam_weight_decay": 0.0001,
+    "adv_clip_max": 5,
+    "batch_size": 64,
+    "beta": 0.005,
+    "cfg": true,
+    "clip_range": 0.0001,
+    "clip_range_gt": 0.0001,
+    "clip_range_lt": 0.0001,
+    "condition_dropout_prob": 0.01,
+    "ema": false,
+    "gradient_accumulation_steps": 1,
+    "learning_rate": 0.0001,
+    "lora_alpha": 64,
+    "lora_path": null,
+    "lora_rank": 32,
+    "max_grad_norm": 1.0,
+    "max_train_steps": 200,
+    "merge_lora_path": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint/results_new/scratch_15k",
+    "num_inner_epochs": 1,
+    "rl_lora_target_modules": [
+      "qkv_proj",
+      "o_proj"
+    ],
+    "timestep_fraction": 1.0,
+    "timestep_shift": 3.0,
+    "use_8bit_adam": false
+  },
+  "use_lora": true
+}
diff --git a/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-10/lora/README.md b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-10/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-10/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-10/lora/adapter_config.json b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-10/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..ec22fc2bfb7e31cdb0828e6cf23b83bbee872655
--- /dev/null
+++ b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-10/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "qkv_proj",
+    "o_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-10/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-10/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..d78f46b077eba7d28eb7396d1b9d6449ca192b71
--- /dev/null
+++ b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-10/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e018de46a9090a5fb16a80bda7c18697cae5a830084c1a672fe97e90865dd58c
+size 37765936
diff --git a/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-100/lora/README.md b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-100/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-100/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-100/lora/adapter_config.json b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-100/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..ec22fc2bfb7e31cdb0828e6cf23b83bbee872655
--- /dev/null
+++ b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-100/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "qkv_proj",
+    "o_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-100/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-100/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..a22513cc049dcac0a121678508f2daf0e3c45d5c
--- /dev/null
+++ b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-100/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bcda783f50161efa0e7e1f18e2a99dc0ad8358b80de50b3db277ddca4739b4a8
+size 37765936
diff --git a/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-110/lora/README.md b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-110/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-110/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-110/lora/adapter_config.json b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-110/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..ec22fc2bfb7e31cdb0828e6cf23b83bbee872655
--- /dev/null
+++ b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-110/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "qkv_proj",
+    "o_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-110/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-110/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..572a4fe1949776301cd9de5c2fb403282aed5d90
--- /dev/null
+++ b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-110/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:96a950e76955e5b380b0c5dc6ffd7b2adbfa87b2575fe696572c4aa31a46e39f
+size 37765936
diff --git a/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-120/lora/README.md b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-120/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-120/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-120/lora/adapter_config.json b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-120/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..ec22fc2bfb7e31cdb0828e6cf23b83bbee872655
--- /dev/null
+++ b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-120/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "qkv_proj",
+    "o_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-120/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-120/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..c809b0a8529ea6f59823a5076ff8247392970fae
--- /dev/null
+++ b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-120/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:34696b0f6b9a4f926b85972a591a296f5afb0452d89a0ba70a10d7609fca896a
+size 37765936
diff --git a/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-130/lora/README.md b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-130/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-130/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-130/lora/adapter_config.json b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-130/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..ec22fc2bfb7e31cdb0828e6cf23b83bbee872655
--- /dev/null
+++ b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-130/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "qkv_proj",
+    "o_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-130/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-130/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..d498a8fe55ee36382790c2eb54929cb14216bd04
--- /dev/null
+++ b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-130/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ffa2d3d010e8d37358621748a5f69264971a4e25d4432b7f9e53ad66230dfac7
+size 37765936
diff --git a/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-140/lora/README.md b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-140/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-140/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-140/lora/adapter_config.json b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-140/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..ec22fc2bfb7e31cdb0828e6cf23b83bbee872655
--- /dev/null
+++ b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-140/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "qkv_proj",
+    "o_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-140/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-140/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..aa02ba4b90ab3decfd8f35057077739384469fd8
--- /dev/null
+++ b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-140/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0071c5abdb93a1353d86856f7044c5fc3c4cf66a59ad633969efcbbc4e39f46d
+size 37765936
diff --git a/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-150/lora/README.md b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-150/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-150/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-150/lora/adapter_config.json b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-150/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..ec22fc2bfb7e31cdb0828e6cf23b83bbee872655
--- /dev/null
+++ b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-150/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "qkv_proj",
+    "o_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-150/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-150/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..2088a353ffd425e355289b54297e911bb89d74da
--- /dev/null
+++ b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-150/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:04c751b05781a4fa82b5170035c3e56ccebd7b0fdbdcb910867ac82c4d22d761
+size 37765936
diff --git a/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-160/lora/README.md b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-160/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-160/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-160/lora/adapter_config.json b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-160/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..ec22fc2bfb7e31cdb0828e6cf23b83bbee872655
--- /dev/null
+++ b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-160/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "qkv_proj",
+    "o_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-160/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-160/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..d123a71816b3c9c61f4b71e979b83d081e921922
--- /dev/null
+++ b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-160/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:468628a76c19e80bd6f1942f85b1b621da45416d3f24829b9e174ce37aebe051
+size 37765936
diff --git a/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-170/lora/README.md b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-170/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-170/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-170/lora/adapter_config.json b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-170/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..ec22fc2bfb7e31cdb0828e6cf23b83bbee872655
--- /dev/null
+++ b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-170/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "qkv_proj",
+    "o_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-170/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-170/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..479779b4a1d8a9c01276855b9214c2c79cb0baac
--- /dev/null
+++ b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-170/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2c315f474b0128b9411f819e65e69186f70c260ecc7baa73ac701141ea440ee7
+size 37765936
diff --git a/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-180/lora/README.md b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-180/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-180/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-180/lora/adapter_config.json b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-180/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..ec22fc2bfb7e31cdb0828e6cf23b83bbee872655
--- /dev/null
+++ b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-180/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "qkv_proj",
+    "o_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-180/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-180/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..b0cfc4f06b1820eee3386d42e531554eaa6cbfbe
--- /dev/null
+++ b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-180/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:18ad0de0f77ce6a7154953ccf3aaf3d489377400b94a61e086748ffd57de8bff
+size 37765936
diff --git a/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-190/lora/README.md b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-190/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-190/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-190/lora/adapter_config.json b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-190/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..ec22fc2bfb7e31cdb0828e6cf23b83bbee872655
--- /dev/null
+++ b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-190/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "qkv_proj",
+    "o_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-190/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-190/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..808378aac163ed1da1d93d77c99f7a20827a61f2
--- /dev/null
+++ b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-190/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ab3015ecf9e9b6d0147c55fdc15b2d6c3f20bff0fd4be0d15353eb20b47462a5
+size 37765936
diff --git a/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-20/lora/README.md b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-20/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-20/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-20/lora/adapter_config.json b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-20/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..ec22fc2bfb7e31cdb0828e6cf23b83bbee872655
--- /dev/null
+++ b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-20/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "qkv_proj",
+    "o_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-20/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-20/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..77a567315028dcd47c10d131a77c0992f56b16a7
--- /dev/null
+++ b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-20/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:80242d41468b8961a723d971a85cfb397cb1b384074de5e62b77ffc6517618c6
+size 37765936
diff --git a/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-30/lora/README.md b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-30/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-30/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-30/lora/adapter_config.json b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-30/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..ec22fc2bfb7e31cdb0828e6cf23b83bbee872655
--- /dev/null
+++ b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-30/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "qkv_proj",
+    "o_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-30/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-30/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..f022c3f4153edaa928c54c2b6bb8d9ac7d8ec046
--- /dev/null
+++ b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-30/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7ee88cf2170aba0959f0ff89380e8abbd2126a4cdc522a88a7b344b28f0db167
+size 37765936
diff --git a/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-40/lora/README.md b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-40/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-40/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-40/lora/adapter_config.json b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-40/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..ec22fc2bfb7e31cdb0828e6cf23b83bbee872655
--- /dev/null
+++ b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-40/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "qkv_proj",
+    "o_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-40/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-40/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..2dfb37b11f9f96fc6aa3a9be587d550acdcbe5f8
--- /dev/null
+++ b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-40/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bb6bf663915fd09e54d704b16e7d319ce1de34a9a50fda2f179ccede105a1b3a
+size 37765936
diff --git a/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-50/lora/README.md b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-50/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-50/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-50/lora/adapter_config.json b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-50/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..ec22fc2bfb7e31cdb0828e6cf23b83bbee872655
--- /dev/null
+++ b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-50/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "qkv_proj",
+    "o_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-50/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-50/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..e3f5e19da42e8732864fbadb47338d429ce7aea0
--- /dev/null
+++ b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-50/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:674c085032f1f55adc618642470f9c18cac4988a74338ca92dc0c14c64225e5f
+size 37765936
diff --git a/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-60/lora/README.md b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-60/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-60/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-60/lora/adapter_config.json b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-60/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..ec22fc2bfb7e31cdb0828e6cf23b83bbee872655
--- /dev/null
+++ b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-60/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "qkv_proj",
+    "o_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-60/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-60/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..f6c576b50a1ff8281df3187bcb4c6385d1e37143
--- /dev/null
+++ b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-60/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:100cd525afbb6f62ddb5cfda68990917dc1bf459348dd59104611d49f1a151db
+size 37765936
diff --git a/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-70/lora/README.md b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-70/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-70/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-70/lora/adapter_config.json b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-70/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..ec22fc2bfb7e31cdb0828e6cf23b83bbee872655
--- /dev/null
+++ b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-70/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "qkv_proj",
+    "o_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-70/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-70/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..cce0309fb1a8050e94f85609aa16dc7ceeb11743
--- /dev/null
+++ b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-70/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c9b4c52e37beee0d1c0de7a443d1681ca051326dacf0663ba728723223633fe0
+size 37765936
diff --git a/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-80/lora/README.md b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-80/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-80/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-80/lora/adapter_config.json b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-80/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..ec22fc2bfb7e31cdb0828e6cf23b83bbee872655
--- /dev/null
+++ b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-80/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "qkv_proj",
+    "o_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-80/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-80/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..6e0fa17e3525d65dff9fe4efda324310a4700db3
--- /dev/null
+++ b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-80/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3b91a6a4d6c712a6ea25e393659c315ddd295b7f5ed17047b925626c21bcdd24
+size 37765936
diff --git a/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-90/lora/README.md b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-90/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-90/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-90/lora/adapter_config.json b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-90/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..ec22fc2bfb7e31cdb0828e6cf23b83bbee872655
--- /dev/null
+++ b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-90/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "qkv_proj",
+    "o_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-90/lora/adapter_model.safetensors b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-90/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..d6a6915eaf19bf33258f10c7237fd31f36d3b7de
--- /dev/null
+++ b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/checkpoints/checkpoint-90/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d98147cd06f09f3d3fa993b8229a07217fcadf04694292de8091feb16dc12c82
+size 37765936
diff --git a/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/environment_summary.txt b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/environment_summary.txt
new file mode 100644
index 0000000000000000000000000000000000000000..be1a1addad40e0e02422e1bafbd3a349806d2c74
--- /dev/null
+++ b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/environment_summary.txt
@@ -0,0 +1,21 @@
+{
+  "environment": {
+    "CUDA_VISIBLE_DEVICES": "0,1,2,3",
+    "HF_HOME": "/home/wenting/.cache/huggingface",
+    "HF_HUB_CACHE": "/home/wenting/.cache/huggingface/hub",
+    "OMNIGEN_CODE_ROOT": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint",
+    "OUTPUT_DIR": "outputs/fix_win/statcm_win8_10_n001_bs64_steps200",
+    "REWARD_MASK_CHANNELS": "0,1",
+    "SAVE_DIR": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/flow_grpo_cxr/logs/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200",
+    "SFT_LORA_PATH": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint/results_new/scratch_15k",
+    "TEST_JSONL": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_test.jsonl",
+    "TRAIN_JSONL": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_train.jsonl",
+    "WANDB_MODE": "online",
+    "WANDB_NAME": "statcm_win8_10_n001_bs64_h20",
+    "WANDB_PROJECT": "fixwin_h20"
+  },
+  "platform": "Linux-5.4.241-1-tlinux4-0017.7-x86_64-with-glibc2.38",
+  "python": "3.10.20 (main, Mar 11 2026, 17:46:40) [GCC 14.3.0]",
+  "repo_root": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/flow_grpo_cxr",
+  "timestamp": "2026-06-08T13:29:03"
+}
diff --git a/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/git_head.txt b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/git_head.txt
new file mode 100644
index 0000000000000000000000000000000000000000..910c3da6325e61af26279bd74233576c2a0530b2
--- /dev/null
+++ b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/git_head.txt
@@ -0,0 +1 @@
+5c0a81d87d64ffd6ca4e4428794155a290835637
diff --git a/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/git_status.txt b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/git_status.txt
new file mode 100644
index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391
diff --git a/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/launch_config.txt b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/launch_config.txt
new file mode 100644
index 0000000000000000000000000000000000000000..95d398b13489d0b6f51ca8e655aefaaff1b3a304
--- /dev/null
+++ b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/launch_config.txt
@@ -0,0 +1,49 @@
+timestamp=20260608_132615
+repo_path=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/flow_grpo_cxr
+config_entry=config/grpo.py:general_radiomics_omnigen_4gpu_kl
+server_profile=
+omnigen_code_root=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint
+sft_lora_path=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint/results_new/scratch_15k
+dataset_root=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset
+train_jsonl=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_train.jsonl
+test_jsonl=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_test.jsonl
+dataset_path_remap_from=/home/wenting/dataset/LIDC_IDRI
+dataset_path_remap_to=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset
+mirp_repo_root=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/mirp
+save_dir=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/flow_grpo_cxr/logs/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200
+output_dir=outputs/fix_win/statcm_win8_10_n001_bs64_steps200
+num_steps=10
+eval_num_steps=50
+noise_level=0.01
+sde_type=cps
+sde_window_enabled=1
+sde_window_start=8
+sde_window_end=10
+sde_window_mode=fixed
+sde_window_count=2
+sde_window_train_only=1
+sde_window_logprob_only=0
+SDE_WINDOW_ENABLED=1
+SDE_WINDOW_START=8
+SDE_WINDOW_END=10
+SDE_WINDOW_MODE=fixed
+SDE_WINDOW_COUNT=2
+guidance_scale=2.5
+img_guidance_scale=2.0
+num_image_per_prompt=4
+train_batch_size=64
+test_batch_size=32
+num_batches_per_epoch=2
+kl_beta=0.005
+learning_rate=1e-4
+max_train_steps=200
+reward_feature_prefixes=stat_,cm_
+reward_base_feature_families=stat,cm
+reward_mask_channels=0,1
+num_processes=4
+wandb_project=fixwin_h20
+wandb_group=manual_img_only
+wandb_name=statcm_win8_10_n001_bs64_h20
+wandb_tags=img-only,scratch30k,manual,noise0.01,bs64
+wandb_mode=online
+wandb_resume=never
diff --git a/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/resolved_config.json b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/resolved_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..ecb3e4ad5836f187ae785e31a44d2a9830eb2a34
--- /dev/null
+++ b/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200/resolved_config.json
@@ -0,0 +1,157 @@
+{
+  "activation_checkpointing": true,
+  "allow_tf32": true,
+  "config_entry": "config/grpo.py:general_radiomics_omnigen_4gpu_kl",
+  "dataset": {
+    "dataset_path_remap_from": "/home/wenting/dataset/LIDC_IDRI",
+    "dataset_path_remap_to": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset",
+    "path_replacements": [
+      [
+        "/home/wenting/dataset/LIDC_IDRI",
+        "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset"
+      ]
+    ],
+    "root": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset",
+    "test_jsonl": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_test.jsonl",
+    "train_jsonl": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_train.jsonl"
+  },
+  "debug": false,
+  "eval_freq": 30,
+  "fsdp_optimizer_offload": false,
+  "logdir": "logs",
+  "mixed_precision": "bf16",
+  "num_checkpoint_limit": 5,
+  "num_epochs": 100000,
+  "per_prompt_stat_tracking": true,
+  "pretrained": {
+    "local_code_root": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint",
+    "model": "Shitao/OmniGen-v1",
+    "revision": "main",
+    "vae_path": null
+  },
+  "prompt_fn": "radiomics_edit",
+  "prompt_fn_kwargs": {},
+  "resolution": 256,
+  "resume_from": null,
+  "reward_fn": {
+    "radiomics": {
+      "base_feature_families": [
+        "stat",
+        "cm"
+      ],
+      "data_root": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset",
+      "dataset_path_remap_from": "/home/wenting/dataset/LIDC_IDRI",
+      "dataset_path_remap_to": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset",
+      "distance_metric": "relative_l1",
+      "feature_prefixes": [
+        "stat_",
+        "cm_"
+      ],
+      "fixed_bin_number": 32,
+      "gt_root": null,
+      "mask_channel": 0,
+      "mask_channels": [
+        0,
+        1
+      ],
+      "mask_root": null,
+      "mirp_repo_root": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/mirp",
+      "missing_score": -10.0,
+      "pixel_spacing_mm": 1.0,
+      "relative_error_clip": 10.0,
+      "reward_scale": 1.0,
+      "roi_name": "lung_roi",
+      "weight": 1.0
+    }
+  },
+  "run_name": "2026.06.08_13.28.58",
+  "sample": {
+    "batch_size": 8,
+    "eval_guidance_scale": 2.5,
+    "eval_img_guidance_scale": 2.0,
+    "eval_num_steps": 50,
+    "global_std": true,
+    "guidance_scale": 2.5,
+    "img_guidance_scale": 2.0,
+    "max_eval_batches": 2,
+    "max_input_image_size": 256,
+    "noise_level": 0.01,
+    "num_batches_per_epoch": 2,
+    "num_image_per_prompt": 4,
+    "num_steps": 10,
+    "same_latent": false,
+    "sde_type": "cps",
+    "sde_window_enabled": true,
+    "sde_window_end": 10,
+    "sde_window_logprob_only": false,
+    "sde_window_mode": "fixed",
+    "sde_window_range": [
+      0,
+      10
+    ],
+    "sde_window_size": 2,
+    "sde_window_start": 8,
+    "sde_window_train_only": true,
+    "test_batch_size": 32,
+    "train_batch_size": 64,
+    "use_img_guidance": true,
+    "use_input_image_size_as_output": false
+  },
+  "save_dir": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/flow_grpo_cxr/logs/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200",
+  "save_freq": 5,
+  "seed": 42,
+  "stage2_extra": {
+    "base_model": "Shitao/OmniGen-v1",
+    "mask_modules_path": null,
+    "omnigen_code_root": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint",
+    "reward_mask_channels": "0,1",
+    "rl_lora_target_modules": [
+      "qkv_proj",
+      "o_proj"
+    ],
+    "save_dir": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/flow_grpo_cxr/logs/radiomics/fix_win/statcm_win8_10_n001_bs64_steps200",
+    "sde_window_enabled": true,
+    "sde_window_end": 10,
+    "sde_window_mode": "fixed",
+    "sde_window_start": 8,
+    "sft_lora_path": null,
+    "test_jsonl": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_test.jsonl",
+    "train_jsonl": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_train.jsonl",
+    "train_timestep_indices": [
+      8,
+      9
+    ]
+  },
+  "train": {
+    "adam_beta1": 0.9,
+    "adam_beta2": 0.999,
+    "adam_epsilon": 1e-08,
+    "adam_weight_decay": 0.0001,
+    "adv_clip_max": 5,
+    "batch_size": 64,
+    "beta": 0.005,
+    "cfg": true,
+    "clip_range": 0.0001,
+    "clip_range_gt": 0.0001,
+    "clip_range_lt": 0.0001,
+    "condition_dropout_prob": 0.01,
+    "ema": false,
+    "gradient_accumulation_steps": 1,
+    "learning_rate": 0.0001,
+    "lora_alpha": 64,
+    "lora_path": null,
+    "lora_rank": 32,
+    "max_grad_norm": 1.0,
+    "max_train_steps": 200,
+    "merge_lora_path": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint/results_new/scratch_15k",
+    "num_inner_epochs": 1,
+    "rl_lora_target_modules": [
+      "qkv_proj",
+      "o_proj"
+    ],
+    "timestep_fraction": 1.0,
+    "timestep_shift": 3.0,
+    "use_8bit_adam": false
+  },
+  "use_lora": true
+}
diff --git a/radiomics/img_only_manual/cm_n001_bs64_steps200/checkpoints/checkpoint-10/lora/README.md b/radiomics/img_only_manual/cm_n001_bs64_steps200/checkpoints/checkpoint-10/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/img_only_manual/cm_n001_bs64_steps200/checkpoints/checkpoint-10/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/img_only_manual/cm_n001_bs64_steps200/checkpoints/checkpoint-10/lora/adapter_config.json b/radiomics/img_only_manual/cm_n001_bs64_steps200/checkpoints/checkpoint-10/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..5caca2d3fe4f842c2d173e65caa2fb5f52a00540
--- /dev/null
+++ b/radiomics/img_only_manual/cm_n001_bs64_steps200/checkpoints/checkpoint-10/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "qkv_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/img_only_manual/cm_n001_bs64_steps200/checkpoints/checkpoint-10/lora/adapter_model.safetensors b/radiomics/img_only_manual/cm_n001_bs64_steps200/checkpoints/checkpoint-10/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..273c355b152c78969763a89556167d7fe8322259
--- /dev/null
+++ b/radiomics/img_only_manual/cm_n001_bs64_steps200/checkpoints/checkpoint-10/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ca57fb93c7db5663c79fdf43616dedcf6ba8998f0c9dbbc75cb193fc121b2742
+size 37765936
diff --git a/radiomics/img_only_manual/cm_n001_bs64_steps200/checkpoints/checkpoint-20/lora/README.md b/radiomics/img_only_manual/cm_n001_bs64_steps200/checkpoints/checkpoint-20/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/img_only_manual/cm_n001_bs64_steps200/checkpoints/checkpoint-20/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/img_only_manual/cm_n001_bs64_steps200/checkpoints/checkpoint-20/lora/adapter_config.json b/radiomics/img_only_manual/cm_n001_bs64_steps200/checkpoints/checkpoint-20/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..5caca2d3fe4f842c2d173e65caa2fb5f52a00540
--- /dev/null
+++ b/radiomics/img_only_manual/cm_n001_bs64_steps200/checkpoints/checkpoint-20/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "qkv_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/img_only_manual/cm_n001_bs64_steps200/checkpoints/checkpoint-20/lora/adapter_model.safetensors b/radiomics/img_only_manual/cm_n001_bs64_steps200/checkpoints/checkpoint-20/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..651c24bebdebbb805cc25765f1caab5187b727b9
--- /dev/null
+++ b/radiomics/img_only_manual/cm_n001_bs64_steps200/checkpoints/checkpoint-20/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:02afc85e42e6132558d906e317c1fe5c323c54769f8b24c7fc6bf44992121890
+size 37765936
diff --git a/radiomics/img_only_manual/cm_n001_bs64_steps200/checkpoints/checkpoint-30/lora/README.md b/radiomics/img_only_manual/cm_n001_bs64_steps200/checkpoints/checkpoint-30/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/img_only_manual/cm_n001_bs64_steps200/checkpoints/checkpoint-30/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/img_only_manual/cm_n001_bs64_steps200/checkpoints/checkpoint-30/lora/adapter_config.json b/radiomics/img_only_manual/cm_n001_bs64_steps200/checkpoints/checkpoint-30/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..5caca2d3fe4f842c2d173e65caa2fb5f52a00540
--- /dev/null
+++ b/radiomics/img_only_manual/cm_n001_bs64_steps200/checkpoints/checkpoint-30/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "qkv_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/img_only_manual/cm_n001_bs64_steps200/checkpoints/checkpoint-30/lora/adapter_model.safetensors b/radiomics/img_only_manual/cm_n001_bs64_steps200/checkpoints/checkpoint-30/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..b5f84114dda6df8bf76dc73ea44ae1f5bb9ddd15
--- /dev/null
+++ b/radiomics/img_only_manual/cm_n001_bs64_steps200/checkpoints/checkpoint-30/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2b812236ff82a0dbdebda9f41d73fb588f6c37ce52d265c72fa7c84b60ea02be
+size 37765936
diff --git a/radiomics/img_only_manual/cm_n001_bs64_steps200/checkpoints/checkpoint-40/lora/README.md b/radiomics/img_only_manual/cm_n001_bs64_steps200/checkpoints/checkpoint-40/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/img_only_manual/cm_n001_bs64_steps200/checkpoints/checkpoint-40/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/img_only_manual/cm_n001_bs64_steps200/checkpoints/checkpoint-40/lora/adapter_config.json b/radiomics/img_only_manual/cm_n001_bs64_steps200/checkpoints/checkpoint-40/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..5caca2d3fe4f842c2d173e65caa2fb5f52a00540
--- /dev/null
+++ b/radiomics/img_only_manual/cm_n001_bs64_steps200/checkpoints/checkpoint-40/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "qkv_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/img_only_manual/cm_n001_bs64_steps200/checkpoints/checkpoint-40/lora/adapter_model.safetensors b/radiomics/img_only_manual/cm_n001_bs64_steps200/checkpoints/checkpoint-40/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..47bd6b1b44e5420fc75f3fc9626905ca3c1b1672
--- /dev/null
+++ b/radiomics/img_only_manual/cm_n001_bs64_steps200/checkpoints/checkpoint-40/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:803ad86656a8e98bb2c077cd67be27f2c42a05db7bcb066cdedcabe5bc8e0e55
+size 37765936
diff --git a/radiomics/img_only_manual/cm_n001_bs64_steps200/checkpoints/checkpoint-50/lora/README.md b/radiomics/img_only_manual/cm_n001_bs64_steps200/checkpoints/checkpoint-50/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/img_only_manual/cm_n001_bs64_steps200/checkpoints/checkpoint-50/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/img_only_manual/cm_n001_bs64_steps200/checkpoints/checkpoint-50/lora/adapter_config.json b/radiomics/img_only_manual/cm_n001_bs64_steps200/checkpoints/checkpoint-50/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..5caca2d3fe4f842c2d173e65caa2fb5f52a00540
--- /dev/null
+++ b/radiomics/img_only_manual/cm_n001_bs64_steps200/checkpoints/checkpoint-50/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "qkv_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/img_only_manual/cm_n001_bs64_steps200/checkpoints/checkpoint-50/lora/adapter_model.safetensors b/radiomics/img_only_manual/cm_n001_bs64_steps200/checkpoints/checkpoint-50/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..723147ca2a72b6bcf9f0d94c0fd1aee869eedd14
--- /dev/null
+++ b/radiomics/img_only_manual/cm_n001_bs64_steps200/checkpoints/checkpoint-50/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:218b6c0b49726d5042ae373041c834bd9129890d671fc7b1d51120e8ced3b52e
+size 37765936
diff --git a/radiomics/img_only_manual/cm_n001_bs64_steps200/checkpoints/checkpoint-60/lora/README.md b/radiomics/img_only_manual/cm_n001_bs64_steps200/checkpoints/checkpoint-60/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/img_only_manual/cm_n001_bs64_steps200/checkpoints/checkpoint-60/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/img_only_manual/cm_n001_bs64_steps200/checkpoints/checkpoint-60/lora/adapter_config.json b/radiomics/img_only_manual/cm_n001_bs64_steps200/checkpoints/checkpoint-60/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..5caca2d3fe4f842c2d173e65caa2fb5f52a00540
--- /dev/null
+++ b/radiomics/img_only_manual/cm_n001_bs64_steps200/checkpoints/checkpoint-60/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "qkv_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/img_only_manual/cm_n001_bs64_steps200/checkpoints/checkpoint-60/lora/adapter_model.safetensors b/radiomics/img_only_manual/cm_n001_bs64_steps200/checkpoints/checkpoint-60/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..139fe74f09d9d2ff718bcb9f344a83e3b2586b34
--- /dev/null
+++ b/radiomics/img_only_manual/cm_n001_bs64_steps200/checkpoints/checkpoint-60/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:62edcc4011e81770a4d663b77e4da330e9d122c49619a1961bd286fbca8b82ad
+size 37765936
diff --git a/radiomics/img_only_manual/cm_n001_bs64_steps200/checkpoints/checkpoint-70/lora/README.md b/radiomics/img_only_manual/cm_n001_bs64_steps200/checkpoints/checkpoint-70/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/img_only_manual/cm_n001_bs64_steps200/checkpoints/checkpoint-70/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/img_only_manual/cm_n001_bs64_steps200/checkpoints/checkpoint-70/lora/adapter_config.json b/radiomics/img_only_manual/cm_n001_bs64_steps200/checkpoints/checkpoint-70/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..5caca2d3fe4f842c2d173e65caa2fb5f52a00540
--- /dev/null
+++ b/radiomics/img_only_manual/cm_n001_bs64_steps200/checkpoints/checkpoint-70/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "qkv_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/img_only_manual/cm_n001_bs64_steps200/checkpoints/checkpoint-70/lora/adapter_model.safetensors b/radiomics/img_only_manual/cm_n001_bs64_steps200/checkpoints/checkpoint-70/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..1cd68f57adb8841cdf84145ca5e36c8d716e137a
--- /dev/null
+++ b/radiomics/img_only_manual/cm_n001_bs64_steps200/checkpoints/checkpoint-70/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e7a787c01fc8043a0ef92eee2b25b7e59ef9e1f74e126dd77a247ffff4159e7f
+size 37765936
diff --git a/radiomics/img_only_manual/cm_n001_bs64_steps200/checkpoints/checkpoint-80/lora/README.md b/radiomics/img_only_manual/cm_n001_bs64_steps200/checkpoints/checkpoint-80/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/img_only_manual/cm_n001_bs64_steps200/checkpoints/checkpoint-80/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/img_only_manual/cm_n001_bs64_steps200/checkpoints/checkpoint-80/lora/adapter_config.json b/radiomics/img_only_manual/cm_n001_bs64_steps200/checkpoints/checkpoint-80/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..5caca2d3fe4f842c2d173e65caa2fb5f52a00540
--- /dev/null
+++ b/radiomics/img_only_manual/cm_n001_bs64_steps200/checkpoints/checkpoint-80/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "qkv_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/img_only_manual/cm_n001_bs64_steps200/checkpoints/checkpoint-80/lora/adapter_model.safetensors b/radiomics/img_only_manual/cm_n001_bs64_steps200/checkpoints/checkpoint-80/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..a3eedc9b4b9c7120b4c47488aa713f8f22df9d5d
--- /dev/null
+++ b/radiomics/img_only_manual/cm_n001_bs64_steps200/checkpoints/checkpoint-80/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2f327ed02ad76e9904ea91859edcaf112ce16402aeac1229597dedbf71488e2e
+size 37765936
diff --git a/radiomics/img_only_manual/cm_n001_bs64_steps200/environment_summary.txt b/radiomics/img_only_manual/cm_n001_bs64_steps200/environment_summary.txt
new file mode 100644
index 0000000000000000000000000000000000000000..7150561019e6226ceafc77a3df770a943d6104ce
--- /dev/null
+++ b/radiomics/img_only_manual/cm_n001_bs64_steps200/environment_summary.txt
@@ -0,0 +1,21 @@
+{
+  "environment": {
+    "CUDA_VISIBLE_DEVICES": "0,1,2,3",
+    "HF_HOME": "/home/wenting/.cache/huggingface",
+    "HF_HUB_CACHE": "/home/wenting/.cache/huggingface/hub",
+    "OMNIGEN_CODE_ROOT": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint",
+    "OUTPUT_DIR": "outputs/img_only_manual/cm_n001_bs64_steps200",
+    "REWARD_MASK_CHANNELS": "0,1",
+    "SAVE_DIR": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/flow_grpo_cxr/logs/radiomics/img_only_manual/cm_n001_bs64_steps200",
+    "SFT_LORA_PATH": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint/results_new/scratch_15k",
+    "TEST_JSONL": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_test.jsonl",
+    "TRAIN_JSONL": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_train.jsonl",
+    "WANDB_MODE": "online",
+    "WANDB_NAME": "cm_n001_bs64_steps200",
+    "WANDB_PROJECT": "flow_grpo_img_only_scratch30k"
+  },
+  "platform": "Linux-5.4.241-1-tlinux4-0017.7-x86_64-with-glibc2.38",
+  "python": "3.10.20 (main, Mar 11 2026, 17:46:40) [GCC 14.3.0]",
+  "repo_root": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/flow_grpo_cxr",
+  "timestamp": "2026-06-06T11:06:09"
+}
diff --git a/radiomics/img_only_manual/cm_n001_bs64_steps200/git_head.txt b/radiomics/img_only_manual/cm_n001_bs64_steps200/git_head.txt
new file mode 100644
index 0000000000000000000000000000000000000000..704f4df26bf92fce31c5ce054a725152eb5f1910
--- /dev/null
+++ b/radiomics/img_only_manual/cm_n001_bs64_steps200/git_head.txt
@@ -0,0 +1 @@
+5b537cea6dba56f0f6308ae8ba43c6f95afaa1bd
diff --git a/radiomics/img_only_manual/cm_n001_bs64_steps200/git_status.txt b/radiomics/img_only_manual/cm_n001_bs64_steps200/git_status.txt
new file mode 100644
index 0000000000000000000000000000000000000000..34f1e98bace6af80c339cab00f83b4c0ebadd27c
--- /dev/null
+++ b/radiomics/img_only_manual/cm_n001_bs64_steps200/git_status.txt
@@ -0,0 +1,16 @@
+ M config/grpo.py
+ M docs/rl_framework_analysis/37_image_only_sweep_runner_guide.md
+ M flow_grpo/radiomics_reward.py
+ M scripts/compute_metrics_for_eval_outputs.py
+ M scripts/debug_preflight_omnigen_env.py
+ M scripts/debug_radiomics_reward_smoke.py
+ M scripts/setup_h20_flow_grpo_env.sh
+ M scripts/single_node/eval_4gpu_scratch15k_image_only.sh
+ M scripts/single_node/grpo_4gpu_scratch15k_image_only.sh
+ M scripts/single_node/metrics_scratch15k_image_only.sh
+ M scripts/single_node/run_img_only_reward_noise_sweep.sh
+ M scripts/train_omnigen.py
+?? analysis_tools/debug_h20_dataset_path_resolution.py
+?? flow_grpo/dataset_paths.py
+?? flow_grpo/server_profiles.py
+?? scripts/single_node/server_profile_defaults.sh
diff --git a/radiomics/img_only_manual/cm_n001_bs64_steps200/launch_config.txt b/radiomics/img_only_manual/cm_n001_bs64_steps200/launch_config.txt
new file mode 100644
index 0000000000000000000000000000000000000000..809ad73512147c629604a3c7f471dbbce28b0791
--- /dev/null
+++ b/radiomics/img_only_manual/cm_n001_bs64_steps200/launch_config.txt
@@ -0,0 +1,37 @@
+timestamp=20260606_110549
+repo_path=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/flow_grpo_cxr
+config_entry=config/grpo.py:general_radiomics_omnigen_4gpu_kl
+server_profile=
+omnigen_code_root=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint
+sft_lora_path=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint/results_new/scratch_15k
+dataset_root=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset
+train_jsonl=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_train.jsonl
+test_jsonl=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_test.jsonl
+dataset_path_remap_from=/home/wenting/dataset/LIDC_IDRI
+dataset_path_remap_to=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset
+mirp_repo_root=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/mirp
+save_dir=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/flow_grpo_cxr/logs/radiomics/img_only_manual/cm_n001_bs64_steps200
+output_dir=outputs/img_only_manual/cm_n001_bs64_steps200
+num_steps=10
+eval_num_steps=50
+noise_level=0.01
+sde_type=cps
+guidance_scale=2.5
+img_guidance_scale=2.0
+num_image_per_prompt=4
+train_batch_size=64
+test_batch_size=32
+num_batches_per_epoch=2
+kl_beta=0.005
+learning_rate=1e-4
+max_train_steps=0
+reward_feature_prefixes=cm_
+reward_base_feature_families=cm
+reward_mask_channels=0,1
+num_processes=4
+wandb_project=flow_grpo_img_only_scratch30k
+wandb_group=manual_img_only
+wandb_name=cm_n001_bs64_steps200
+wandb_tags=img-only,scratch30k,manual,noise0.01,bs64
+wandb_mode=online
+wandb_resume=never
diff --git a/radiomics/img_only_manual/cm_n001_bs64_steps200/resolved_config.json b/radiomics/img_only_manual/cm_n001_bs64_steps200/resolved_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..cc5480f5fb46a305b048ead8f6bf3d0f944624d1
--- /dev/null
+++ b/radiomics/img_only_manual/cm_n001_bs64_steps200/resolved_config.json
@@ -0,0 +1,141 @@
+{
+  "activation_checkpointing": true,
+  "allow_tf32": true,
+  "config_entry": "config/grpo.py:general_radiomics_omnigen_4gpu_kl",
+  "dataset": {
+    "dataset_path_remap_from": "/home/wenting/dataset/LIDC_IDRI",
+    "dataset_path_remap_to": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset",
+    "path_replacements": [
+      [
+        "/home/wenting/dataset/LIDC_IDRI",
+        "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset"
+      ]
+    ],
+    "root": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset",
+    "test_jsonl": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_test.jsonl",
+    "train_jsonl": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_train.jsonl"
+  },
+  "debug": false,
+  "eval_freq": 30,
+  "fsdp_optimizer_offload": false,
+  "logdir": "logs",
+  "mixed_precision": "bf16",
+  "num_checkpoint_limit": 5,
+  "num_epochs": 100000,
+  "per_prompt_stat_tracking": true,
+  "pretrained": {
+    "local_code_root": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint",
+    "model": "Shitao/OmniGen-v1",
+    "revision": "main",
+    "vae_path": null
+  },
+  "prompt_fn": "radiomics_edit",
+  "prompt_fn_kwargs": {},
+  "resolution": 256,
+  "resume_from": null,
+  "reward_fn": {
+    "radiomics": {
+      "base_feature_families": [
+        "cm"
+      ],
+      "data_root": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset",
+      "dataset_path_remap_from": "/home/wenting/dataset/LIDC_IDRI",
+      "dataset_path_remap_to": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset",
+      "distance_metric": "relative_l1",
+      "feature_prefixes": [
+        "cm_"
+      ],
+      "fixed_bin_number": 32,
+      "gt_root": null,
+      "mask_channel": 0,
+      "mask_channels": [
+        0,
+        1
+      ],
+      "mask_root": null,
+      "mirp_repo_root": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/mirp",
+      "missing_score": -10.0,
+      "pixel_spacing_mm": 1.0,
+      "relative_error_clip": 10.0,
+      "reward_scale": 1.0,
+      "roi_name": "lung_roi",
+      "weight": 1.0
+    }
+  },
+  "run_name": "2026.06.06_11.06.05",
+  "sample": {
+    "batch_size": 8,
+    "eval_guidance_scale": 2.5,
+    "eval_img_guidance_scale": 2.0,
+    "eval_num_steps": 50,
+    "global_std": true,
+    "guidance_scale": 2.5,
+    "img_guidance_scale": 2.0,
+    "max_eval_batches": 2,
+    "max_input_image_size": 256,
+    "noise_level": 0.01,
+    "num_batches_per_epoch": 2,
+    "num_image_per_prompt": 4,
+    "num_steps": 10,
+    "same_latent": false,
+    "sde_type": "cps",
+    "sde_window_range": [
+      0,
+      10
+    ],
+    "sde_window_size": 2,
+    "test_batch_size": 32,
+    "train_batch_size": 64,
+    "use_img_guidance": true,
+    "use_input_image_size_as_output": false
+  },
+  "save_dir": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/flow_grpo_cxr/logs/radiomics/img_only_manual/cm_n001_bs64_steps200",
+  "save_freq": 5,
+  "seed": 42,
+  "stage2_extra": {
+    "base_model": "Shitao/OmniGen-v1",
+    "mask_modules_path": null,
+    "omnigen_code_root": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint",
+    "reward_mask_channels": "0,1",
+    "rl_lora_target_modules": [
+      "qkv_proj",
+      "o_proj"
+    ],
+    "save_dir": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/flow_grpo_cxr/logs/radiomics/img_only_manual/cm_n001_bs64_steps200",
+    "sft_lora_path": null,
+    "test_jsonl": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_test.jsonl",
+    "train_jsonl": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_train.jsonl"
+  },
+  "train": {
+    "adam_beta1": 0.9,
+    "adam_beta2": 0.999,
+    "adam_epsilon": 1e-08,
+    "adam_weight_decay": 0.0001,
+    "adv_clip_max": 5,
+    "batch_size": 64,
+    "beta": 0.005,
+    "cfg": true,
+    "clip_range": 0.0001,
+    "clip_range_gt": 0.0001,
+    "clip_range_lt": 0.0001,
+    "condition_dropout_prob": 0.01,
+    "ema": false,
+    "gradient_accumulation_steps": 1,
+    "learning_rate": 0.0001,
+    "lora_alpha": 64,
+    "lora_path": null,
+    "lora_rank": 32,
+    "max_grad_norm": 1.0,
+    "max_train_steps": 0,
+    "merge_lora_path": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint/results_new/scratch_15k",
+    "num_inner_epochs": 1,
+    "rl_lora_target_modules": [
+      "qkv_proj",
+      "o_proj"
+    ],
+    "timestep_fraction": 1.0,
+    "timestep_shift": 3.0,
+    "use_8bit_adam": false
+  },
+  "use_lora": true
+}
diff --git a/radiomics/img_only_manual/stat_n001_bs64_steps200/checkpoints/checkpoint-10/lora/README.md b/radiomics/img_only_manual/stat_n001_bs64_steps200/checkpoints/checkpoint-10/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/img_only_manual/stat_n001_bs64_steps200/checkpoints/checkpoint-10/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/img_only_manual/stat_n001_bs64_steps200/checkpoints/checkpoint-10/lora/adapter_config.json b/radiomics/img_only_manual/stat_n001_bs64_steps200/checkpoints/checkpoint-10/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..ec22fc2bfb7e31cdb0828e6cf23b83bbee872655
--- /dev/null
+++ b/radiomics/img_only_manual/stat_n001_bs64_steps200/checkpoints/checkpoint-10/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "qkv_proj",
+    "o_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/img_only_manual/stat_n001_bs64_steps200/checkpoints/checkpoint-10/lora/adapter_model.safetensors b/radiomics/img_only_manual/stat_n001_bs64_steps200/checkpoints/checkpoint-10/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..8d403c6ea6a27511a21eb67404cd2bb0122176ca
--- /dev/null
+++ b/radiomics/img_only_manual/stat_n001_bs64_steps200/checkpoints/checkpoint-10/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:03e1a6ea905895106a56f667bbb677d8c2b8d1aee3dae8776aa4b2b8c3d691fb
+size 37765936
diff --git a/radiomics/img_only_manual/stat_n001_bs64_steps200/checkpoints/checkpoint-20/lora/README.md b/radiomics/img_only_manual/stat_n001_bs64_steps200/checkpoints/checkpoint-20/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/img_only_manual/stat_n001_bs64_steps200/checkpoints/checkpoint-20/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/img_only_manual/stat_n001_bs64_steps200/checkpoints/checkpoint-20/lora/adapter_config.json b/radiomics/img_only_manual/stat_n001_bs64_steps200/checkpoints/checkpoint-20/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..ec22fc2bfb7e31cdb0828e6cf23b83bbee872655
--- /dev/null
+++ b/radiomics/img_only_manual/stat_n001_bs64_steps200/checkpoints/checkpoint-20/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "qkv_proj",
+    "o_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/img_only_manual/stat_n001_bs64_steps200/checkpoints/checkpoint-20/lora/adapter_model.safetensors b/radiomics/img_only_manual/stat_n001_bs64_steps200/checkpoints/checkpoint-20/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..49d9a87889b71ba170f91a425027951cabfbc3ab
--- /dev/null
+++ b/radiomics/img_only_manual/stat_n001_bs64_steps200/checkpoints/checkpoint-20/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:86d7a6fa8cfc3e7c876dac6e77b80d68f20ead5fdc2f26f7b55732d561ed0e33
+size 37765936
diff --git a/radiomics/img_only_manual/stat_n001_bs64_steps200/checkpoints/checkpoint-30/lora/README.md b/radiomics/img_only_manual/stat_n001_bs64_steps200/checkpoints/checkpoint-30/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/img_only_manual/stat_n001_bs64_steps200/checkpoints/checkpoint-30/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/img_only_manual/stat_n001_bs64_steps200/checkpoints/checkpoint-30/lora/adapter_config.json b/radiomics/img_only_manual/stat_n001_bs64_steps200/checkpoints/checkpoint-30/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..ec22fc2bfb7e31cdb0828e6cf23b83bbee872655
--- /dev/null
+++ b/radiomics/img_only_manual/stat_n001_bs64_steps200/checkpoints/checkpoint-30/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "qkv_proj",
+    "o_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/img_only_manual/stat_n001_bs64_steps200/checkpoints/checkpoint-30/lora/adapter_model.safetensors b/radiomics/img_only_manual/stat_n001_bs64_steps200/checkpoints/checkpoint-30/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..2b0105e82d7cad3bd6e672dc2ab7f0bc0675e904
--- /dev/null
+++ b/radiomics/img_only_manual/stat_n001_bs64_steps200/checkpoints/checkpoint-30/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0c478772959a91c0d07368d545da1093b55f31514308c7daec649017e1601ec8
+size 37765936
diff --git a/radiomics/img_only_manual/stat_n001_bs64_steps200/checkpoints/checkpoint-40/lora/README.md b/radiomics/img_only_manual/stat_n001_bs64_steps200/checkpoints/checkpoint-40/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/img_only_manual/stat_n001_bs64_steps200/checkpoints/checkpoint-40/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/img_only_manual/stat_n001_bs64_steps200/checkpoints/checkpoint-40/lora/adapter_config.json b/radiomics/img_only_manual/stat_n001_bs64_steps200/checkpoints/checkpoint-40/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..ec22fc2bfb7e31cdb0828e6cf23b83bbee872655
--- /dev/null
+++ b/radiomics/img_only_manual/stat_n001_bs64_steps200/checkpoints/checkpoint-40/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "qkv_proj",
+    "o_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/img_only_manual/stat_n001_bs64_steps200/checkpoints/checkpoint-40/lora/adapter_model.safetensors b/radiomics/img_only_manual/stat_n001_bs64_steps200/checkpoints/checkpoint-40/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..b0732f120c1ab6adeb8b34059e48b6eb410ee822
--- /dev/null
+++ b/radiomics/img_only_manual/stat_n001_bs64_steps200/checkpoints/checkpoint-40/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b6c1b773c416a9dbafb1be5d9324b69b49326f10bbcdb8064aa13ccf043eca2e
+size 37765936
diff --git a/radiomics/img_only_manual/stat_n001_bs64_steps200/checkpoints/checkpoint-50/lora/README.md b/radiomics/img_only_manual/stat_n001_bs64_steps200/checkpoints/checkpoint-50/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/img_only_manual/stat_n001_bs64_steps200/checkpoints/checkpoint-50/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/img_only_manual/stat_n001_bs64_steps200/checkpoints/checkpoint-50/lora/adapter_config.json b/radiomics/img_only_manual/stat_n001_bs64_steps200/checkpoints/checkpoint-50/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..ec22fc2bfb7e31cdb0828e6cf23b83bbee872655
--- /dev/null
+++ b/radiomics/img_only_manual/stat_n001_bs64_steps200/checkpoints/checkpoint-50/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "qkv_proj",
+    "o_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/img_only_manual/stat_n001_bs64_steps200/checkpoints/checkpoint-50/lora/adapter_model.safetensors b/radiomics/img_only_manual/stat_n001_bs64_steps200/checkpoints/checkpoint-50/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..085b03af42dbe75ef09744c742089ad5b20c302a
--- /dev/null
+++ b/radiomics/img_only_manual/stat_n001_bs64_steps200/checkpoints/checkpoint-50/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:575fe61f10476abe49f501db605c1a3ea3771799f8e6750455755a381c945f9c
+size 37765936
diff --git a/radiomics/img_only_manual/stat_n001_bs64_steps200/checkpoints/checkpoint-60/lora/README.md b/radiomics/img_only_manual/stat_n001_bs64_steps200/checkpoints/checkpoint-60/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/img_only_manual/stat_n001_bs64_steps200/checkpoints/checkpoint-60/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/img_only_manual/stat_n001_bs64_steps200/checkpoints/checkpoint-60/lora/adapter_config.json b/radiomics/img_only_manual/stat_n001_bs64_steps200/checkpoints/checkpoint-60/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..ec22fc2bfb7e31cdb0828e6cf23b83bbee872655
--- /dev/null
+++ b/radiomics/img_only_manual/stat_n001_bs64_steps200/checkpoints/checkpoint-60/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "qkv_proj",
+    "o_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/img_only_manual/stat_n001_bs64_steps200/checkpoints/checkpoint-60/lora/adapter_model.safetensors b/radiomics/img_only_manual/stat_n001_bs64_steps200/checkpoints/checkpoint-60/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..3535cff18e1af404f27a7607f8332b089d0eb2ea
--- /dev/null
+++ b/radiomics/img_only_manual/stat_n001_bs64_steps200/checkpoints/checkpoint-60/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e4c5c678b3beebfd0d3e305bcf9daebda09253d78e22e4b589a20580f9a97da2
+size 37765936
diff --git a/radiomics/img_only_manual/stat_n001_bs64_steps200/checkpoints/checkpoint-70/lora/README.md b/radiomics/img_only_manual/stat_n001_bs64_steps200/checkpoints/checkpoint-70/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/img_only_manual/stat_n001_bs64_steps200/checkpoints/checkpoint-70/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/img_only_manual/stat_n001_bs64_steps200/checkpoints/checkpoint-70/lora/adapter_config.json b/radiomics/img_only_manual/stat_n001_bs64_steps200/checkpoints/checkpoint-70/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..ec22fc2bfb7e31cdb0828e6cf23b83bbee872655
--- /dev/null
+++ b/radiomics/img_only_manual/stat_n001_bs64_steps200/checkpoints/checkpoint-70/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "qkv_proj",
+    "o_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/img_only_manual/stat_n001_bs64_steps200/checkpoints/checkpoint-70/lora/adapter_model.safetensors b/radiomics/img_only_manual/stat_n001_bs64_steps200/checkpoints/checkpoint-70/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..0187a32c4de5cdcee01588f9dc57b360c0de0839
--- /dev/null
+++ b/radiomics/img_only_manual/stat_n001_bs64_steps200/checkpoints/checkpoint-70/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dacc32e0cb69007e342b7a39507c87c7b7de9ef11174a6546c2affc01cc52414
+size 37765936
diff --git a/radiomics/img_only_manual/stat_n001_bs64_steps200/environment_summary.txt b/radiomics/img_only_manual/stat_n001_bs64_steps200/environment_summary.txt
new file mode 100644
index 0000000000000000000000000000000000000000..8c2c2e597087235a8447f07a7ee1e1263503683b
--- /dev/null
+++ b/radiomics/img_only_manual/stat_n001_bs64_steps200/environment_summary.txt
@@ -0,0 +1,21 @@
+{
+  "environment": {
+    "CUDA_VISIBLE_DEVICES": "0,1,2,3",
+    "HF_HOME": "/home/wenting/.cache/huggingface",
+    "HF_HUB_CACHE": "/home/wenting/.cache/huggingface/hub",
+    "OMNIGEN_CODE_ROOT": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint",
+    "OUTPUT_DIR": "outputs/img_only_manual/stat_n001_bs64_steps200",
+    "REWARD_MASK_CHANNELS": "0,1",
+    "SAVE_DIR": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/flow_grpo_cxr/logs/radiomics/img_only_manual/stat_n001_bs64_steps200",
+    "SFT_LORA_PATH": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint/results_new/scratch_15k",
+    "TEST_JSONL": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_test.jsonl",
+    "TRAIN_JSONL": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_train.jsonl",
+    "WANDB_MODE": "online",
+    "WANDB_NAME": "stat_n001_bs64_steps200",
+    "WANDB_PROJECT": "flow_grpo_img_only_scratch30k"
+  },
+  "platform": "Linux-5.4.241-1-tlinux4-0017.7-x86_64-with-glibc2.38",
+  "python": "3.10.20 (main, Mar 11 2026, 17:46:40) [GCC 14.3.0]",
+  "repo_root": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/flow_grpo_cxr",
+  "timestamp": "2026-06-06T11:23:32"
+}
diff --git a/radiomics/img_only_manual/stat_n001_bs64_steps200/git_head.txt b/radiomics/img_only_manual/stat_n001_bs64_steps200/git_head.txt
new file mode 100644
index 0000000000000000000000000000000000000000..704f4df26bf92fce31c5ce054a725152eb5f1910
--- /dev/null
+++ b/radiomics/img_only_manual/stat_n001_bs64_steps200/git_head.txt
@@ -0,0 +1 @@
+5b537cea6dba56f0f6308ae8ba43c6f95afaa1bd
diff --git a/radiomics/img_only_manual/stat_n001_bs64_steps200/git_status.txt b/radiomics/img_only_manual/stat_n001_bs64_steps200/git_status.txt
new file mode 100644
index 0000000000000000000000000000000000000000..34f1e98bace6af80c339cab00f83b4c0ebadd27c
--- /dev/null
+++ b/radiomics/img_only_manual/stat_n001_bs64_steps200/git_status.txt
@@ -0,0 +1,16 @@
+ M config/grpo.py
+ M docs/rl_framework_analysis/37_image_only_sweep_runner_guide.md
+ M flow_grpo/radiomics_reward.py
+ M scripts/compute_metrics_for_eval_outputs.py
+ M scripts/debug_preflight_omnigen_env.py
+ M scripts/debug_radiomics_reward_smoke.py
+ M scripts/setup_h20_flow_grpo_env.sh
+ M scripts/single_node/eval_4gpu_scratch15k_image_only.sh
+ M scripts/single_node/grpo_4gpu_scratch15k_image_only.sh
+ M scripts/single_node/metrics_scratch15k_image_only.sh
+ M scripts/single_node/run_img_only_reward_noise_sweep.sh
+ M scripts/train_omnigen.py
+?? analysis_tools/debug_h20_dataset_path_resolution.py
+?? flow_grpo/dataset_paths.py
+?? flow_grpo/server_profiles.py
+?? scripts/single_node/server_profile_defaults.sh
diff --git a/radiomics/img_only_manual/stat_n001_bs64_steps200/launch_config.txt b/radiomics/img_only_manual/stat_n001_bs64_steps200/launch_config.txt
new file mode 100644
index 0000000000000000000000000000000000000000..6e414a5949ee2ecb0bb902d76f6ca3444212e26f
--- /dev/null
+++ b/radiomics/img_only_manual/stat_n001_bs64_steps200/launch_config.txt
@@ -0,0 +1,37 @@
+timestamp=20260606_112037
+repo_path=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/flow_grpo_cxr
+config_entry=config/grpo.py:general_radiomics_omnigen_4gpu_kl
+server_profile=
+omnigen_code_root=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint
+sft_lora_path=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint/results_new/scratch_15k
+dataset_root=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset
+train_jsonl=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_train.jsonl
+test_jsonl=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_test.jsonl
+dataset_path_remap_from=/home/wenting/dataset/LIDC_IDRI
+dataset_path_remap_to=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset
+mirp_repo_root=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/mirp
+save_dir=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/flow_grpo_cxr/logs/radiomics/img_only_manual/stat_n001_bs64_steps200
+output_dir=outputs/img_only_manual/stat_n001_bs64_steps200
+num_steps=10
+eval_num_steps=50
+noise_level=0.01
+sde_type=cps
+guidance_scale=2.5
+img_guidance_scale=2.0
+num_image_per_prompt=4
+train_batch_size=64
+test_batch_size=32
+num_batches_per_epoch=2
+kl_beta=0.005
+learning_rate=1e-4
+max_train_steps=0
+reward_feature_prefixes=stat_
+reward_base_feature_families=stat
+reward_mask_channels=0,1
+num_processes=4
+wandb_project=flow_grpo_img_only_scratch30k
+wandb_group=manual_img_only
+wandb_name=stat_n001_bs64_steps200
+wandb_tags=img-only,scratch30k,manual,noise0.01,bs64
+wandb_mode=online
+wandb_resume=never
diff --git a/radiomics/img_only_manual/stat_n001_bs64_steps200/resolved_config.json b/radiomics/img_only_manual/stat_n001_bs64_steps200/resolved_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..f7ff79353415c257bf744a2b9c1b59ec62577485
--- /dev/null
+++ b/radiomics/img_only_manual/stat_n001_bs64_steps200/resolved_config.json
@@ -0,0 +1,141 @@
+{
+  "activation_checkpointing": true,
+  "allow_tf32": true,
+  "config_entry": "config/grpo.py:general_radiomics_omnigen_4gpu_kl",
+  "dataset": {
+    "dataset_path_remap_from": "/home/wenting/dataset/LIDC_IDRI",
+    "dataset_path_remap_to": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset",
+    "path_replacements": [
+      [
+        "/home/wenting/dataset/LIDC_IDRI",
+        "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset"
+      ]
+    ],
+    "root": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset",
+    "test_jsonl": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_test.jsonl",
+    "train_jsonl": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_train.jsonl"
+  },
+  "debug": false,
+  "eval_freq": 30,
+  "fsdp_optimizer_offload": false,
+  "logdir": "logs",
+  "mixed_precision": "bf16",
+  "num_checkpoint_limit": 5,
+  "num_epochs": 100000,
+  "per_prompt_stat_tracking": true,
+  "pretrained": {
+    "local_code_root": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint",
+    "model": "Shitao/OmniGen-v1",
+    "revision": "main",
+    "vae_path": null
+  },
+  "prompt_fn": "radiomics_edit",
+  "prompt_fn_kwargs": {},
+  "resolution": 256,
+  "resume_from": null,
+  "reward_fn": {
+    "radiomics": {
+      "base_feature_families": [
+        "stat"
+      ],
+      "data_root": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset",
+      "dataset_path_remap_from": "/home/wenting/dataset/LIDC_IDRI",
+      "dataset_path_remap_to": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset",
+      "distance_metric": "relative_l1",
+      "feature_prefixes": [
+        "stat_"
+      ],
+      "fixed_bin_number": 32,
+      "gt_root": null,
+      "mask_channel": 0,
+      "mask_channels": [
+        0,
+        1
+      ],
+      "mask_root": null,
+      "mirp_repo_root": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/mirp",
+      "missing_score": -10.0,
+      "pixel_spacing_mm": 1.0,
+      "relative_error_clip": 10.0,
+      "reward_scale": 1.0,
+      "roi_name": "lung_roi",
+      "weight": 1.0
+    }
+  },
+  "run_name": "2026.06.06_11.23.28",
+  "sample": {
+    "batch_size": 8,
+    "eval_guidance_scale": 2.5,
+    "eval_img_guidance_scale": 2.0,
+    "eval_num_steps": 50,
+    "global_std": true,
+    "guidance_scale": 2.5,
+    "img_guidance_scale": 2.0,
+    "max_eval_batches": 2,
+    "max_input_image_size": 256,
+    "noise_level": 0.01,
+    "num_batches_per_epoch": 2,
+    "num_image_per_prompt": 4,
+    "num_steps": 10,
+    "same_latent": false,
+    "sde_type": "cps",
+    "sde_window_range": [
+      0,
+      10
+    ],
+    "sde_window_size": 2,
+    "test_batch_size": 32,
+    "train_batch_size": 64,
+    "use_img_guidance": true,
+    "use_input_image_size_as_output": false
+  },
+  "save_dir": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/flow_grpo_cxr/logs/radiomics/img_only_manual/stat_n001_bs64_steps200",
+  "save_freq": 5,
+  "seed": 42,
+  "stage2_extra": {
+    "base_model": "Shitao/OmniGen-v1",
+    "mask_modules_path": null,
+    "omnigen_code_root": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint",
+    "reward_mask_channels": "0,1",
+    "rl_lora_target_modules": [
+      "qkv_proj",
+      "o_proj"
+    ],
+    "save_dir": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/flow_grpo_cxr/logs/radiomics/img_only_manual/stat_n001_bs64_steps200",
+    "sft_lora_path": null,
+    "test_jsonl": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_test.jsonl",
+    "train_jsonl": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_train.jsonl"
+  },
+  "train": {
+    "adam_beta1": 0.9,
+    "adam_beta2": 0.999,
+    "adam_epsilon": 1e-08,
+    "adam_weight_decay": 0.0001,
+    "adv_clip_max": 5,
+    "batch_size": 64,
+    "beta": 0.005,
+    "cfg": true,
+    "clip_range": 0.0001,
+    "clip_range_gt": 0.0001,
+    "clip_range_lt": 0.0001,
+    "condition_dropout_prob": 0.01,
+    "ema": false,
+    "gradient_accumulation_steps": 1,
+    "learning_rate": 0.0001,
+    "lora_alpha": 64,
+    "lora_path": null,
+    "lora_rank": 32,
+    "max_grad_norm": 1.0,
+    "max_train_steps": 0,
+    "merge_lora_path": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint/results_new/scratch_15k",
+    "num_inner_epochs": 1,
+    "rl_lora_target_modules": [
+      "qkv_proj",
+      "o_proj"
+    ],
+    "timestep_fraction": 1.0,
+    "timestep_shift": 3.0,
+    "use_8bit_adam": false
+  },
+  "use_lora": true
+}
diff --git a/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-10/lora/README.md b/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-10/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-10/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-10/lora/adapter_config.json b/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-10/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..ec22fc2bfb7e31cdb0828e6cf23b83bbee872655
--- /dev/null
+++ b/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-10/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "qkv_proj",
+    "o_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-10/lora/adapter_model.safetensors b/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-10/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..7e3d74adbe2c5d8068d56eb8616d3615ceba91a1
--- /dev/null
+++ b/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-10/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6b5d7bbb4d4854113ee4c73470882da51730cb4411e771e8280d508b97c79a67
+size 37765936
diff --git a/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-20/lora/README.md b/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-20/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-20/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-20/lora/adapter_config.json b/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-20/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..ec22fc2bfb7e31cdb0828e6cf23b83bbee872655
--- /dev/null
+++ b/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-20/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "qkv_proj",
+    "o_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-20/lora/adapter_model.safetensors b/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-20/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..77d336314bad0b8d98f986c8f546d55e2e608fe5
--- /dev/null
+++ b/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-20/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:336f0bf728683fbd18e8c525ead7523e206d147f9e9f4dba927404beae1434dc
+size 37765936
diff --git a/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-30/lora/README.md b/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-30/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-30/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-30/lora/adapter_config.json b/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-30/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..ec22fc2bfb7e31cdb0828e6cf23b83bbee872655
--- /dev/null
+++ b/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-30/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "qkv_proj",
+    "o_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-30/lora/adapter_model.safetensors b/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-30/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..ab3bd8c4055d689b6727d55f0247b15f65bdd0ae
--- /dev/null
+++ b/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-30/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ef9ce157764d343b28026d37d9725f0c38102fd7e0592e0677e4d5ac8d6b4bd0
+size 37765936
diff --git a/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-40/lora/README.md b/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-40/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-40/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-40/lora/adapter_config.json b/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-40/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..ec22fc2bfb7e31cdb0828e6cf23b83bbee872655
--- /dev/null
+++ b/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-40/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "qkv_proj",
+    "o_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-40/lora/adapter_model.safetensors b/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-40/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..4c2160cdad4c3124bd4bc754b1ec40c45e4100a1
--- /dev/null
+++ b/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-40/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:28f1ca2b8b4c886ba6a192c214329082b48e2db0c572d4bca808f5d2f8172825
+size 37765936
diff --git a/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-50/lora/README.md b/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-50/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-50/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-50/lora/adapter_config.json b/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-50/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..ec22fc2bfb7e31cdb0828e6cf23b83bbee872655
--- /dev/null
+++ b/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-50/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "qkv_proj",
+    "o_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-50/lora/adapter_model.safetensors b/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-50/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..e5d0f093fe87f10de8f4845b2f10b801243a92c0
--- /dev/null
+++ b/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-50/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:720b9a9fb4998579cf489db4276a500e9a543efabf1023b8264f6c5ef6326756
+size 37765936
diff --git a/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-60/lora/README.md b/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-60/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-60/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-60/lora/adapter_config.json b/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-60/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..ec22fc2bfb7e31cdb0828e6cf23b83bbee872655
--- /dev/null
+++ b/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-60/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "qkv_proj",
+    "o_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-60/lora/adapter_model.safetensors b/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-60/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..dd391063d2824b7c350e00daf9e5deeb8de5b73e
--- /dev/null
+++ b/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-60/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:be84405719101c82a8d16941d243d87f521bb8d6c074a63f2be55e28e4f9cfa1
+size 37765936
diff --git a/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-70/lora/README.md b/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-70/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-70/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-70/lora/adapter_config.json b/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-70/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..ec22fc2bfb7e31cdb0828e6cf23b83bbee872655
--- /dev/null
+++ b/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-70/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "qkv_proj",
+    "o_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-70/lora/adapter_model.safetensors b/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-70/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..02c2ade9204b0b74f1a5823a06caf6d1574af703
--- /dev/null
+++ b/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-70/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:43062f802d0f0edbb24c4ee10f159f2503feab543db2ec3a2df3da6a84b2b6bf
+size 37765936
diff --git a/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-80/lora/README.md b/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-80/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-80/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-80/lora/adapter_config.json b/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-80/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..ec22fc2bfb7e31cdb0828e6cf23b83bbee872655
--- /dev/null
+++ b/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-80/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "qkv_proj",
+    "o_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-80/lora/adapter_model.safetensors b/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-80/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..b7ed38e8508fea599598573ce40205d40337ebd8
--- /dev/null
+++ b/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-80/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:afd07c6b80ca9ceffa398ef384f499556b878fb0439ea45e76fe98795bb508ff
+size 37765936
diff --git a/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-90/lora/README.md b/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-90/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-90/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-90/lora/adapter_config.json b/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-90/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..ec22fc2bfb7e31cdb0828e6cf23b83bbee872655
--- /dev/null
+++ b/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-90/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "qkv_proj",
+    "o_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-90/lora/adapter_model.safetensors b/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-90/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..c58d3bce4a161c610f7d933895d685934122db26
--- /dev/null
+++ b/radiomics/img_only_manual/stat_n005_bs64_steps200/checkpoints/checkpoint-90/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ca18cb3f1294077641fbd347880ad8a34e4d660d047a3b0932e1a206878b2fc5
+size 37765936
diff --git a/radiomics/img_only_manual/stat_n005_bs64_steps200/environment_summary.txt b/radiomics/img_only_manual/stat_n005_bs64_steps200/environment_summary.txt
new file mode 100644
index 0000000000000000000000000000000000000000..b7cd27eb055a4d0c11f898d6e19c3c7cc0e0e786
--- /dev/null
+++ b/radiomics/img_only_manual/stat_n005_bs64_steps200/environment_summary.txt
@@ -0,0 +1,21 @@
+{
+  "environment": {
+    "CUDA_VISIBLE_DEVICES": "4,5,6,7",
+    "HF_HOME": "/home/wenting/.cache/huggingface",
+    "HF_HUB_CACHE": "/home/wenting/.cache/huggingface/hub",
+    "OMNIGEN_CODE_ROOT": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint",
+    "OUTPUT_DIR": "outputs/img_only_manual/stat_n005_bs64_steps200",
+    "REWARD_MASK_CHANNELS": "0,1",
+    "SAVE_DIR": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/flow_grpo_cxr/logs/radiomics/img_only_manual/stat_n005_bs64_steps200",
+    "SFT_LORA_PATH": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint/results_new/scratch_15k",
+    "TEST_JSONL": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_test.jsonl",
+    "TRAIN_JSONL": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_train.jsonl",
+    "WANDB_MODE": "online",
+    "WANDB_NAME": "stat_n005_bs64_steps200",
+    "WANDB_PROJECT": "flow_grpo_img_only_scratch30k"
+  },
+  "platform": "Linux-5.4.241-1-tlinux4-0017.7-x86_64-with-glibc2.38",
+  "python": "3.10.20 (main, Mar 11 2026, 17:46:40) [GCC 14.3.0]",
+  "repo_root": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/flow_grpo_cxr",
+  "timestamp": "2026-06-06T11:14:07"
+}
diff --git a/radiomics/img_only_manual/stat_n005_bs64_steps200/git_head.txt b/radiomics/img_only_manual/stat_n005_bs64_steps200/git_head.txt
new file mode 100644
index 0000000000000000000000000000000000000000..704f4df26bf92fce31c5ce054a725152eb5f1910
--- /dev/null
+++ b/radiomics/img_only_manual/stat_n005_bs64_steps200/git_head.txt
@@ -0,0 +1 @@
+5b537cea6dba56f0f6308ae8ba43c6f95afaa1bd
diff --git a/radiomics/img_only_manual/stat_n005_bs64_steps200/git_status.txt b/radiomics/img_only_manual/stat_n005_bs64_steps200/git_status.txt
new file mode 100644
index 0000000000000000000000000000000000000000..34f1e98bace6af80c339cab00f83b4c0ebadd27c
--- /dev/null
+++ b/radiomics/img_only_manual/stat_n005_bs64_steps200/git_status.txt
@@ -0,0 +1,16 @@
+ M config/grpo.py
+ M docs/rl_framework_analysis/37_image_only_sweep_runner_guide.md
+ M flow_grpo/radiomics_reward.py
+ M scripts/compute_metrics_for_eval_outputs.py
+ M scripts/debug_preflight_omnigen_env.py
+ M scripts/debug_radiomics_reward_smoke.py
+ M scripts/setup_h20_flow_grpo_env.sh
+ M scripts/single_node/eval_4gpu_scratch15k_image_only.sh
+ M scripts/single_node/grpo_4gpu_scratch15k_image_only.sh
+ M scripts/single_node/metrics_scratch15k_image_only.sh
+ M scripts/single_node/run_img_only_reward_noise_sweep.sh
+ M scripts/train_omnigen.py
+?? analysis_tools/debug_h20_dataset_path_resolution.py
+?? flow_grpo/dataset_paths.py
+?? flow_grpo/server_profiles.py
+?? scripts/single_node/server_profile_defaults.sh
diff --git a/radiomics/img_only_manual/stat_n005_bs64_steps200/launch_config.txt b/radiomics/img_only_manual/stat_n005_bs64_steps200/launch_config.txt
new file mode 100644
index 0000000000000000000000000000000000000000..0f16f73dc24cd19e79644ca8b781729d6007251b
--- /dev/null
+++ b/radiomics/img_only_manual/stat_n005_bs64_steps200/launch_config.txt
@@ -0,0 +1,37 @@
+timestamp=20260606_111341
+repo_path=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/flow_grpo_cxr
+config_entry=config/grpo.py:general_radiomics_omnigen_4gpu_kl
+server_profile=
+omnigen_code_root=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint
+sft_lora_path=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint/results_new/scratch_15k
+dataset_root=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset
+train_jsonl=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_train.jsonl
+test_jsonl=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_test.jsonl
+dataset_path_remap_from=/home/wenting/dataset/LIDC_IDRI
+dataset_path_remap_to=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset
+mirp_repo_root=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/mirp
+save_dir=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/flow_grpo_cxr/logs/radiomics/img_only_manual/stat_n005_bs64_steps200
+output_dir=outputs/img_only_manual/stat_n005_bs64_steps200
+num_steps=10
+eval_num_steps=50
+noise_level=0.005
+sde_type=cps
+guidance_scale=2.5
+img_guidance_scale=2.0
+num_image_per_prompt=4
+train_batch_size=64
+test_batch_size=32
+num_batches_per_epoch=2
+kl_beta=0.005
+learning_rate=1e-4
+max_train_steps=0
+reward_feature_prefixes=stat_
+reward_base_feature_families=stat
+reward_mask_channels=0,1
+num_processes=4
+wandb_project=flow_grpo_img_only_scratch30k
+wandb_group=manual_img_only
+wandb_name=stat_n005_bs64_steps200
+wandb_tags=img-only,scratch30k,manual,noise0.005,bs64
+wandb_mode=online
+wandb_resume=never
diff --git a/radiomics/img_only_manual/stat_n005_bs64_steps200/resolved_config.json b/radiomics/img_only_manual/stat_n005_bs64_steps200/resolved_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..51322886f27c1f1fa6213dc2387f233f3191440c
--- /dev/null
+++ b/radiomics/img_only_manual/stat_n005_bs64_steps200/resolved_config.json
@@ -0,0 +1,141 @@
+{
+  "activation_checkpointing": true,
+  "allow_tf32": true,
+  "config_entry": "config/grpo.py:general_radiomics_omnigen_4gpu_kl",
+  "dataset": {
+    "dataset_path_remap_from": "/home/wenting/dataset/LIDC_IDRI",
+    "dataset_path_remap_to": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset",
+    "path_replacements": [
+      [
+        "/home/wenting/dataset/LIDC_IDRI",
+        "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset"
+      ]
+    ],
+    "root": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset",
+    "test_jsonl": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_test.jsonl",
+    "train_jsonl": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_train.jsonl"
+  },
+  "debug": false,
+  "eval_freq": 30,
+  "fsdp_optimizer_offload": false,
+  "logdir": "logs",
+  "mixed_precision": "bf16",
+  "num_checkpoint_limit": 5,
+  "num_epochs": 100000,
+  "per_prompt_stat_tracking": true,
+  "pretrained": {
+    "local_code_root": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint",
+    "model": "Shitao/OmniGen-v1",
+    "revision": "main",
+    "vae_path": null
+  },
+  "prompt_fn": "radiomics_edit",
+  "prompt_fn_kwargs": {},
+  "resolution": 256,
+  "resume_from": null,
+  "reward_fn": {
+    "radiomics": {
+      "base_feature_families": [
+        "stat"
+      ],
+      "data_root": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset",
+      "dataset_path_remap_from": "/home/wenting/dataset/LIDC_IDRI",
+      "dataset_path_remap_to": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset",
+      "distance_metric": "relative_l1",
+      "feature_prefixes": [
+        "stat_"
+      ],
+      "fixed_bin_number": 32,
+      "gt_root": null,
+      "mask_channel": 0,
+      "mask_channels": [
+        0,
+        1
+      ],
+      "mask_root": null,
+      "mirp_repo_root": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/mirp",
+      "missing_score": -10.0,
+      "pixel_spacing_mm": 1.0,
+      "relative_error_clip": 10.0,
+      "reward_scale": 1.0,
+      "roi_name": "lung_roi",
+      "weight": 1.0
+    }
+  },
+  "run_name": "2026.06.06_11.14.02",
+  "sample": {
+    "batch_size": 8,
+    "eval_guidance_scale": 2.5,
+    "eval_img_guidance_scale": 2.0,
+    "eval_num_steps": 50,
+    "global_std": true,
+    "guidance_scale": 2.5,
+    "img_guidance_scale": 2.0,
+    "max_eval_batches": 2,
+    "max_input_image_size": 256,
+    "noise_level": 0.005,
+    "num_batches_per_epoch": 2,
+    "num_image_per_prompt": 4,
+    "num_steps": 10,
+    "same_latent": false,
+    "sde_type": "cps",
+    "sde_window_range": [
+      0,
+      10
+    ],
+    "sde_window_size": 2,
+    "test_batch_size": 32,
+    "train_batch_size": 64,
+    "use_img_guidance": true,
+    "use_input_image_size_as_output": false
+  },
+  "save_dir": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/flow_grpo_cxr/logs/radiomics/img_only_manual/stat_n005_bs64_steps200",
+  "save_freq": 5,
+  "seed": 42,
+  "stage2_extra": {
+    "base_model": "Shitao/OmniGen-v1",
+    "mask_modules_path": null,
+    "omnigen_code_root": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint",
+    "reward_mask_channels": "0,1",
+    "rl_lora_target_modules": [
+      "qkv_proj",
+      "o_proj"
+    ],
+    "save_dir": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/flow_grpo_cxr/logs/radiomics/img_only_manual/stat_n005_bs64_steps200",
+    "sft_lora_path": null,
+    "test_jsonl": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_test.jsonl",
+    "train_jsonl": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_train.jsonl"
+  },
+  "train": {
+    "adam_beta1": 0.9,
+    "adam_beta2": 0.999,
+    "adam_epsilon": 1e-08,
+    "adam_weight_decay": 0.0001,
+    "adv_clip_max": 5,
+    "batch_size": 64,
+    "beta": 0.005,
+    "cfg": true,
+    "clip_range": 0.0001,
+    "clip_range_gt": 0.0001,
+    "clip_range_lt": 0.0001,
+    "condition_dropout_prob": 0.01,
+    "ema": false,
+    "gradient_accumulation_steps": 1,
+    "learning_rate": 0.0001,
+    "lora_alpha": 64,
+    "lora_path": null,
+    "lora_rank": 32,
+    "max_grad_norm": 1.0,
+    "max_train_steps": 0,
+    "merge_lora_path": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint/results_new/scratch_15k",
+    "num_inner_epochs": 1,
+    "rl_lora_target_modules": [
+      "qkv_proj",
+      "o_proj"
+    ],
+    "timestep_fraction": 1.0,
+    "timestep_shift": 3.0,
+    "use_8bit_adam": false
+  },
+  "use_lora": true
+}
diff --git a/radiomics/img_only_manual/statcm_n001_bs32_steps200/environment_summary.txt b/radiomics/img_only_manual/statcm_n001_bs32_steps200/environment_summary.txt
new file mode 100644
index 0000000000000000000000000000000000000000..46d9f82c3d094050670c6a9a38d45979ecba4ce1
--- /dev/null
+++ b/radiomics/img_only_manual/statcm_n001_bs32_steps200/environment_summary.txt
@@ -0,0 +1,21 @@
+{
+  "environment": {
+    "CUDA_VISIBLE_DEVICES": "0,1,2,3",
+    "HF_HOME": "/home/wenting/.cache/huggingface",
+    "HF_HUB_CACHE": "/home/wenting/.cache/huggingface/hub",
+    "OMNIGEN_CODE_ROOT": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint",
+    "OUTPUT_DIR": "outputs/img_only_manual/statcm_n001_bs32_steps200",
+    "REWARD_MASK_CHANNELS": "0,1",
+    "SAVE_DIR": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/flow_grpo_cxr/logs/radiomics/img_only_manual/statcm_n001_bs32_steps200",
+    "SFT_LORA_PATH": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint/results_new/scratch_15k",
+    "TEST_JSONL": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_test.jsonl",
+    "TRAIN_JSONL": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_train.jsonl",
+    "WANDB_MODE": "online",
+    "WANDB_NAME": "statcm_n001_bs32_steps200",
+    "WANDB_PROJECT": "flow_grpo_img_only_scratch30k"
+  },
+  "platform": "Linux-5.4.241-1-tlinux4-0017.7-x86_64-with-glibc2.38",
+  "python": "3.10.20 (main, Mar 11 2026, 17:46:40) [GCC 14.3.0]",
+  "repo_root": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/flow_grpo_cxr",
+  "timestamp": "2026-06-06T10:15:43"
+}
diff --git a/radiomics/img_only_manual/statcm_n001_bs32_steps200/git_head.txt b/radiomics/img_only_manual/statcm_n001_bs32_steps200/git_head.txt
new file mode 100644
index 0000000000000000000000000000000000000000..704f4df26bf92fce31c5ce054a725152eb5f1910
--- /dev/null
+++ b/radiomics/img_only_manual/statcm_n001_bs32_steps200/git_head.txt
@@ -0,0 +1 @@
+5b537cea6dba56f0f6308ae8ba43c6f95afaa1bd
diff --git a/radiomics/img_only_manual/statcm_n001_bs32_steps200/git_status.txt b/radiomics/img_only_manual/statcm_n001_bs32_steps200/git_status.txt
new file mode 100644
index 0000000000000000000000000000000000000000..34f1e98bace6af80c339cab00f83b4c0ebadd27c
--- /dev/null
+++ b/radiomics/img_only_manual/statcm_n001_bs32_steps200/git_status.txt
@@ -0,0 +1,16 @@
+ M config/grpo.py
+ M docs/rl_framework_analysis/37_image_only_sweep_runner_guide.md
+ M flow_grpo/radiomics_reward.py
+ M scripts/compute_metrics_for_eval_outputs.py
+ M scripts/debug_preflight_omnigen_env.py
+ M scripts/debug_radiomics_reward_smoke.py
+ M scripts/setup_h20_flow_grpo_env.sh
+ M scripts/single_node/eval_4gpu_scratch15k_image_only.sh
+ M scripts/single_node/grpo_4gpu_scratch15k_image_only.sh
+ M scripts/single_node/metrics_scratch15k_image_only.sh
+ M scripts/single_node/run_img_only_reward_noise_sweep.sh
+ M scripts/train_omnigen.py
+?? analysis_tools/debug_h20_dataset_path_resolution.py
+?? flow_grpo/dataset_paths.py
+?? flow_grpo/server_profiles.py
+?? scripts/single_node/server_profile_defaults.sh
diff --git a/radiomics/img_only_manual/statcm_n001_bs32_steps200/launch_config.txt b/radiomics/img_only_manual/statcm_n001_bs32_steps200/launch_config.txt
new file mode 100644
index 0000000000000000000000000000000000000000..b9de95d572b01016cb37ef5706d9925306253865
--- /dev/null
+++ b/radiomics/img_only_manual/statcm_n001_bs32_steps200/launch_config.txt
@@ -0,0 +1,37 @@
+timestamp=20260606_101522
+repo_path=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/flow_grpo_cxr
+config_entry=config/grpo.py:general_radiomics_omnigen_4gpu_kl
+server_profile=
+omnigen_code_root=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint
+sft_lora_path=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint/results_new/scratch_15k
+dataset_root=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset
+train_jsonl=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_train.jsonl
+test_jsonl=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_test.jsonl
+dataset_path_remap_from=/home/wenting/dataset/LIDC_IDRI
+dataset_path_remap_to=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset
+mirp_repo_root=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/mirp
+save_dir=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/flow_grpo_cxr/logs/radiomics/img_only_manual/statcm_n001_bs32_steps200
+output_dir=outputs/img_only_manual/statcm_n001_bs32_steps200
+num_steps=10
+eval_num_steps=50
+noise_level=0.01
+sde_type=cps
+guidance_scale=2.5
+img_guidance_scale=2.0
+num_image_per_prompt=4
+train_batch_size=32
+test_batch_size=32
+num_batches_per_epoch=2
+kl_beta=0.005
+learning_rate=1e-4
+max_train_steps=0
+reward_feature_prefixes=stat_,cm_
+reward_base_feature_families=stat,cm
+reward_mask_channels=0,1
+num_processes=4
+wandb_project=flow_grpo_img_only_scratch30k
+wandb_group=manual_img_only
+wandb_name=statcm_n001_bs32_steps200
+wandb_tags=img-only,scratch30k,manual,noise0.01,bs32
+wandb_mode=online
+wandb_resume=never
diff --git a/radiomics/img_only_manual/statcm_n001_bs32_steps200/resolved_config.json b/radiomics/img_only_manual/statcm_n001_bs32_steps200/resolved_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1b183750a942c554f9f1b7ed3f83d07628618998
--- /dev/null
+++ b/radiomics/img_only_manual/statcm_n001_bs32_steps200/resolved_config.json
@@ -0,0 +1,143 @@
+{
+  "activation_checkpointing": true,
+  "allow_tf32": true,
+  "config_entry": "config/grpo.py:general_radiomics_omnigen_4gpu_kl",
+  "dataset": {
+    "dataset_path_remap_from": "/home/wenting/dataset/LIDC_IDRI",
+    "dataset_path_remap_to": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset",
+    "path_replacements": [
+      [
+        "/home/wenting/dataset/LIDC_IDRI",
+        "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset"
+      ]
+    ],
+    "root": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset",
+    "test_jsonl": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_test.jsonl",
+    "train_jsonl": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_train.jsonl"
+  },
+  "debug": false,
+  "eval_freq": 30,
+  "fsdp_optimizer_offload": false,
+  "logdir": "logs",
+  "mixed_precision": "bf16",
+  "num_checkpoint_limit": 5,
+  "num_epochs": 100000,
+  "per_prompt_stat_tracking": true,
+  "pretrained": {
+    "local_code_root": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint",
+    "model": "Shitao/OmniGen-v1",
+    "revision": "main",
+    "vae_path": null
+  },
+  "prompt_fn": "radiomics_edit",
+  "prompt_fn_kwargs": {},
+  "resolution": 256,
+  "resume_from": null,
+  "reward_fn": {
+    "radiomics": {
+      "base_feature_families": [
+        "stat",
+        "cm"
+      ],
+      "data_root": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset",
+      "dataset_path_remap_from": "/home/wenting/dataset/LIDC_IDRI",
+      "dataset_path_remap_to": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset",
+      "distance_metric": "relative_l1",
+      "feature_prefixes": [
+        "stat_",
+        "cm_"
+      ],
+      "fixed_bin_number": 32,
+      "gt_root": null,
+      "mask_channel": 0,
+      "mask_channels": [
+        0,
+        1
+      ],
+      "mask_root": null,
+      "mirp_repo_root": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/mirp",
+      "missing_score": -10.0,
+      "pixel_spacing_mm": 1.0,
+      "relative_error_clip": 10.0,
+      "reward_scale": 1.0,
+      "roi_name": "lung_roi",
+      "weight": 1.0
+    }
+  },
+  "run_name": "2026.06.06_10.15.38",
+  "sample": {
+    "batch_size": 8,
+    "eval_guidance_scale": 2.5,
+    "eval_img_guidance_scale": 2.0,
+    "eval_num_steps": 50,
+    "global_std": true,
+    "guidance_scale": 2.5,
+    "img_guidance_scale": 2.0,
+    "max_eval_batches": 2,
+    "max_input_image_size": 256,
+    "noise_level": 0.01,
+    "num_batches_per_epoch": 2,
+    "num_image_per_prompt": 4,
+    "num_steps": 10,
+    "same_latent": false,
+    "sde_type": "cps",
+    "sde_window_range": [
+      0,
+      10
+    ],
+    "sde_window_size": 2,
+    "test_batch_size": 32,
+    "train_batch_size": 32,
+    "use_img_guidance": true,
+    "use_input_image_size_as_output": false
+  },
+  "save_dir": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/flow_grpo_cxr/logs/radiomics/img_only_manual/statcm_n001_bs32_steps200",
+  "save_freq": 5,
+  "seed": 42,
+  "stage2_extra": {
+    "base_model": "Shitao/OmniGen-v1",
+    "mask_modules_path": null,
+    "omnigen_code_root": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint",
+    "reward_mask_channels": "0,1",
+    "rl_lora_target_modules": [
+      "qkv_proj",
+      "o_proj"
+    ],
+    "save_dir": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/flow_grpo_cxr/logs/radiomics/img_only_manual/statcm_n001_bs32_steps200",
+    "sft_lora_path": null,
+    "test_jsonl": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_test.jsonl",
+    "train_jsonl": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_train.jsonl"
+  },
+  "train": {
+    "adam_beta1": 0.9,
+    "adam_beta2": 0.999,
+    "adam_epsilon": 1e-08,
+    "adam_weight_decay": 0.0001,
+    "adv_clip_max": 5,
+    "batch_size": 32,
+    "beta": 0.005,
+    "cfg": true,
+    "clip_range": 0.0001,
+    "clip_range_gt": 0.0001,
+    "clip_range_lt": 0.0001,
+    "condition_dropout_prob": 0.01,
+    "ema": false,
+    "gradient_accumulation_steps": 1,
+    "learning_rate": 0.0001,
+    "lora_alpha": 64,
+    "lora_path": null,
+    "lora_rank": 32,
+    "max_grad_norm": 1.0,
+    "max_train_steps": 0,
+    "merge_lora_path": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint/results_new/scratch_15k",
+    "num_inner_epochs": 1,
+    "rl_lora_target_modules": [
+      "qkv_proj",
+      "o_proj"
+    ],
+    "timestep_fraction": 1.0,
+    "timestep_shift": 3.0,
+    "use_8bit_adam": false
+  },
+  "use_lora": true
+}
diff --git a/radiomics/img_only_manual/statcm_n005_bs64_steps200/checkpoints/checkpoint-10/lora/README.md b/radiomics/img_only_manual/statcm_n005_bs64_steps200/checkpoints/checkpoint-10/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/img_only_manual/statcm_n005_bs64_steps200/checkpoints/checkpoint-10/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/img_only_manual/statcm_n005_bs64_steps200/checkpoints/checkpoint-10/lora/adapter_config.json b/radiomics/img_only_manual/statcm_n005_bs64_steps200/checkpoints/checkpoint-10/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..5caca2d3fe4f842c2d173e65caa2fb5f52a00540
--- /dev/null
+++ b/radiomics/img_only_manual/statcm_n005_bs64_steps200/checkpoints/checkpoint-10/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "qkv_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/img_only_manual/statcm_n005_bs64_steps200/checkpoints/checkpoint-10/lora/adapter_model.safetensors b/radiomics/img_only_manual/statcm_n005_bs64_steps200/checkpoints/checkpoint-10/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..f70d024b48ce6d4d8fc6cbfebef3dd69fb31aaee
--- /dev/null
+++ b/radiomics/img_only_manual/statcm_n005_bs64_steps200/checkpoints/checkpoint-10/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a72c630be71021507cc3cea1d212195e93c65fbd189fe7343109a06cdc0fe5c1
+size 37765936
diff --git a/radiomics/img_only_manual/statcm_n005_bs64_steps200/checkpoints/checkpoint-20/lora/README.md b/radiomics/img_only_manual/statcm_n005_bs64_steps200/checkpoints/checkpoint-20/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/img_only_manual/statcm_n005_bs64_steps200/checkpoints/checkpoint-20/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/img_only_manual/statcm_n005_bs64_steps200/checkpoints/checkpoint-20/lora/adapter_config.json b/radiomics/img_only_manual/statcm_n005_bs64_steps200/checkpoints/checkpoint-20/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..5caca2d3fe4f842c2d173e65caa2fb5f52a00540
--- /dev/null
+++ b/radiomics/img_only_manual/statcm_n005_bs64_steps200/checkpoints/checkpoint-20/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "qkv_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/img_only_manual/statcm_n005_bs64_steps200/checkpoints/checkpoint-20/lora/adapter_model.safetensors b/radiomics/img_only_manual/statcm_n005_bs64_steps200/checkpoints/checkpoint-20/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..d2302aced162ee38d01ff9919b31903d244c8389
--- /dev/null
+++ b/radiomics/img_only_manual/statcm_n005_bs64_steps200/checkpoints/checkpoint-20/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3ac17754da00be147df7a5658360902cf75d7073d08427c6a8a5d357c856b24a
+size 37765936
diff --git a/radiomics/img_only_manual/statcm_n005_bs64_steps200/checkpoints/checkpoint-30/lora/README.md b/radiomics/img_only_manual/statcm_n005_bs64_steps200/checkpoints/checkpoint-30/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/img_only_manual/statcm_n005_bs64_steps200/checkpoints/checkpoint-30/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/img_only_manual/statcm_n005_bs64_steps200/checkpoints/checkpoint-30/lora/adapter_config.json b/radiomics/img_only_manual/statcm_n005_bs64_steps200/checkpoints/checkpoint-30/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..5caca2d3fe4f842c2d173e65caa2fb5f52a00540
--- /dev/null
+++ b/radiomics/img_only_manual/statcm_n005_bs64_steps200/checkpoints/checkpoint-30/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "qkv_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/img_only_manual/statcm_n005_bs64_steps200/checkpoints/checkpoint-30/lora/adapter_model.safetensors b/radiomics/img_only_manual/statcm_n005_bs64_steps200/checkpoints/checkpoint-30/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..b85c0c60c709614ee8cedbe53720106d76578a46
--- /dev/null
+++ b/radiomics/img_only_manual/statcm_n005_bs64_steps200/checkpoints/checkpoint-30/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d9d30d2e75c262a574e1544fca9555217eec4c95945b7f0bd330124b0abce465
+size 37765936
diff --git a/radiomics/img_only_manual/statcm_n005_bs64_steps200/checkpoints/checkpoint-40/lora/README.md b/radiomics/img_only_manual/statcm_n005_bs64_steps200/checkpoints/checkpoint-40/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/img_only_manual/statcm_n005_bs64_steps200/checkpoints/checkpoint-40/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/img_only_manual/statcm_n005_bs64_steps200/checkpoints/checkpoint-40/lora/adapter_config.json b/radiomics/img_only_manual/statcm_n005_bs64_steps200/checkpoints/checkpoint-40/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..5caca2d3fe4f842c2d173e65caa2fb5f52a00540
--- /dev/null
+++ b/radiomics/img_only_manual/statcm_n005_bs64_steps200/checkpoints/checkpoint-40/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "qkv_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/img_only_manual/statcm_n005_bs64_steps200/checkpoints/checkpoint-40/lora/adapter_model.safetensors b/radiomics/img_only_manual/statcm_n005_bs64_steps200/checkpoints/checkpoint-40/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..c4f1c16ff4c8ef53e84e2b8ee4ad107e6e866409
--- /dev/null
+++ b/radiomics/img_only_manual/statcm_n005_bs64_steps200/checkpoints/checkpoint-40/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cfb9cf71b932608f6f7a8a3525630d7f39f34b486a5b03db965600741ebc310f
+size 37765936
diff --git a/radiomics/img_only_manual/statcm_n005_bs64_steps200/checkpoints/checkpoint-50/lora/README.md b/radiomics/img_only_manual/statcm_n005_bs64_steps200/checkpoints/checkpoint-50/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/img_only_manual/statcm_n005_bs64_steps200/checkpoints/checkpoint-50/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/img_only_manual/statcm_n005_bs64_steps200/checkpoints/checkpoint-50/lora/adapter_config.json b/radiomics/img_only_manual/statcm_n005_bs64_steps200/checkpoints/checkpoint-50/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..5caca2d3fe4f842c2d173e65caa2fb5f52a00540
--- /dev/null
+++ b/radiomics/img_only_manual/statcm_n005_bs64_steps200/checkpoints/checkpoint-50/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "qkv_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/img_only_manual/statcm_n005_bs64_steps200/checkpoints/checkpoint-50/lora/adapter_model.safetensors b/radiomics/img_only_manual/statcm_n005_bs64_steps200/checkpoints/checkpoint-50/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..f9c4fc7de622dec36799ec9421ff92adc463c4b1
--- /dev/null
+++ b/radiomics/img_only_manual/statcm_n005_bs64_steps200/checkpoints/checkpoint-50/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ddd704484467b1ba0983b7c32ce79eeb326dac3ba1fa7d496e5481a1777d13bd
+size 37765936
diff --git a/radiomics/img_only_manual/statcm_n005_bs64_steps200/checkpoints/checkpoint-60/lora/README.md b/radiomics/img_only_manual/statcm_n005_bs64_steps200/checkpoints/checkpoint-60/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/img_only_manual/statcm_n005_bs64_steps200/checkpoints/checkpoint-60/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/img_only_manual/statcm_n005_bs64_steps200/checkpoints/checkpoint-60/lora/adapter_config.json b/radiomics/img_only_manual/statcm_n005_bs64_steps200/checkpoints/checkpoint-60/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..5caca2d3fe4f842c2d173e65caa2fb5f52a00540
--- /dev/null
+++ b/radiomics/img_only_manual/statcm_n005_bs64_steps200/checkpoints/checkpoint-60/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "qkv_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/img_only_manual/statcm_n005_bs64_steps200/checkpoints/checkpoint-60/lora/adapter_model.safetensors b/radiomics/img_only_manual/statcm_n005_bs64_steps200/checkpoints/checkpoint-60/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..b7152c8c0519cbf8bd6ab6abad08d8526c5ccfdd
--- /dev/null
+++ b/radiomics/img_only_manual/statcm_n005_bs64_steps200/checkpoints/checkpoint-60/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0178cc47c7647231c3d2819b2e6496a9c7b6149d0432c001ef0bf184d01b6440
+size 37765936
diff --git a/radiomics/img_only_manual/statcm_n005_bs64_steps200/checkpoints/checkpoint-70/lora/README.md b/radiomics/img_only_manual/statcm_n005_bs64_steps200/checkpoints/checkpoint-70/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/img_only_manual/statcm_n005_bs64_steps200/checkpoints/checkpoint-70/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/img_only_manual/statcm_n005_bs64_steps200/checkpoints/checkpoint-70/lora/adapter_config.json b/radiomics/img_only_manual/statcm_n005_bs64_steps200/checkpoints/checkpoint-70/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..5caca2d3fe4f842c2d173e65caa2fb5f52a00540
--- /dev/null
+++ b/radiomics/img_only_manual/statcm_n005_bs64_steps200/checkpoints/checkpoint-70/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "qkv_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/img_only_manual/statcm_n005_bs64_steps200/checkpoints/checkpoint-70/lora/adapter_model.safetensors b/radiomics/img_only_manual/statcm_n005_bs64_steps200/checkpoints/checkpoint-70/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..d3b977ce7daab6f26fe07aae9381635739e61cd9
--- /dev/null
+++ b/radiomics/img_only_manual/statcm_n005_bs64_steps200/checkpoints/checkpoint-70/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1add71dd5c341c8b641f66c4aab358d3e42f4f234ebc70b8790a66f78e89b92c
+size 37765936
diff --git a/radiomics/img_only_manual/statcm_n005_bs64_steps200/checkpoints/checkpoint-80/lora/README.md b/radiomics/img_only_manual/statcm_n005_bs64_steps200/checkpoints/checkpoint-80/lora/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86c9eedd3aa734a240cfacc101cd21b927d397cd
--- /dev/null
+++ b/radiomics/img_only_manual/statcm_n005_bs64_steps200/checkpoints/checkpoint-80/lora/README.md
@@ -0,0 +1,203 @@
+---
+library_name: peft
+tags:
+- lora
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1
\ No newline at end of file
diff --git a/radiomics/img_only_manual/statcm_n005_bs64_steps200/checkpoints/checkpoint-80/lora/adapter_config.json b/radiomics/img_only_manual/statcm_n005_bs64_steps200/checkpoints/checkpoint-80/lora/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..5caca2d3fe4f842c2d173e65caa2fb5f52a00540
--- /dev/null
+++ b/radiomics/img_only_manual/statcm_n005_bs64_steps200/checkpoints/checkpoint-80/lora/adapter_config.json
@@ -0,0 +1,44 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "OmniGen",
+    "parent_library": "OmniGen.model"
+  },
+  "base_model_name_or_path": null,
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "qkv_proj"
+  ],
+  "target_parameters": null,
+  "task_type": null,
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/radiomics/img_only_manual/statcm_n005_bs64_steps200/checkpoints/checkpoint-80/lora/adapter_model.safetensors b/radiomics/img_only_manual/statcm_n005_bs64_steps200/checkpoints/checkpoint-80/lora/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..d067a4d5f27c1d43d4ca530c560825e50b3ac853
--- /dev/null
+++ b/radiomics/img_only_manual/statcm_n005_bs64_steps200/checkpoints/checkpoint-80/lora/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c8441bafcfa546bae5cc9df54b1a819c971b2280af0bf09a913989656ab0857b
+size 37765936
diff --git a/radiomics/img_only_manual/statcm_n005_bs64_steps200/environment_summary.txt b/radiomics/img_only_manual/statcm_n005_bs64_steps200/environment_summary.txt
new file mode 100644
index 0000000000000000000000000000000000000000..0b5fd100a28671383000ab9399e8e20216c89c31
--- /dev/null
+++ b/radiomics/img_only_manual/statcm_n005_bs64_steps200/environment_summary.txt
@@ -0,0 +1,21 @@
+{
+  "environment": {
+    "CUDA_VISIBLE_DEVICES": "4,5,6,7",
+    "HF_HOME": "/home/wenting/.cache/huggingface",
+    "HF_HUB_CACHE": "/home/wenting/.cache/huggingface/hub",
+    "OMNIGEN_CODE_ROOT": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint",
+    "OUTPUT_DIR": "outputs/img_only_manual/statcm_n005_bs64_steps200",
+    "REWARD_MASK_CHANNELS": "0,1",
+    "SAVE_DIR": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/flow_grpo_cxr/logs/radiomics/img_only_manual/statcm_n005_bs64_steps200",
+    "SFT_LORA_PATH": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint/results_new/scratch_15k",
+    "TEST_JSONL": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_test.jsonl",
+    "TRAIN_JSONL": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_train.jsonl",
+    "WANDB_MODE": "online",
+    "WANDB_NAME": "statcm_n005_bs64_steps200",
+    "WANDB_PROJECT": "flow_grpo_img_only_scratch30k"
+  },
+  "platform": "Linux-5.4.241-1-tlinux4-0017.7-x86_64-with-glibc2.38",
+  "python": "3.10.20 (main, Mar 11 2026, 17:46:40) [GCC 14.3.0]",
+  "repo_root": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/flow_grpo_cxr",
+  "timestamp": "2026-06-06T11:27:52"
+}
diff --git a/radiomics/img_only_manual/statcm_n005_bs64_steps200/git_head.txt b/radiomics/img_only_manual/statcm_n005_bs64_steps200/git_head.txt
new file mode 100644
index 0000000000000000000000000000000000000000..704f4df26bf92fce31c5ce054a725152eb5f1910
--- /dev/null
+++ b/radiomics/img_only_manual/statcm_n005_bs64_steps200/git_head.txt
@@ -0,0 +1 @@
+5b537cea6dba56f0f6308ae8ba43c6f95afaa1bd
diff --git a/radiomics/img_only_manual/statcm_n005_bs64_steps200/git_status.txt b/radiomics/img_only_manual/statcm_n005_bs64_steps200/git_status.txt
new file mode 100644
index 0000000000000000000000000000000000000000..34f1e98bace6af80c339cab00f83b4c0ebadd27c
--- /dev/null
+++ b/radiomics/img_only_manual/statcm_n005_bs64_steps200/git_status.txt
@@ -0,0 +1,16 @@
+ M config/grpo.py
+ M docs/rl_framework_analysis/37_image_only_sweep_runner_guide.md
+ M flow_grpo/radiomics_reward.py
+ M scripts/compute_metrics_for_eval_outputs.py
+ M scripts/debug_preflight_omnigen_env.py
+ M scripts/debug_radiomics_reward_smoke.py
+ M scripts/setup_h20_flow_grpo_env.sh
+ M scripts/single_node/eval_4gpu_scratch15k_image_only.sh
+ M scripts/single_node/grpo_4gpu_scratch15k_image_only.sh
+ M scripts/single_node/metrics_scratch15k_image_only.sh
+ M scripts/single_node/run_img_only_reward_noise_sweep.sh
+ M scripts/train_omnigen.py
+?? analysis_tools/debug_h20_dataset_path_resolution.py
+?? flow_grpo/dataset_paths.py
+?? flow_grpo/server_profiles.py
+?? scripts/single_node/server_profile_defaults.sh
diff --git a/radiomics/img_only_manual/statcm_n005_bs64_steps200/launch_config.txt b/radiomics/img_only_manual/statcm_n005_bs64_steps200/launch_config.txt
new file mode 100644
index 0000000000000000000000000000000000000000..0fa1c2ce68aa052edb4b98fc4012756b3ea2a979
--- /dev/null
+++ b/radiomics/img_only_manual/statcm_n005_bs64_steps200/launch_config.txt
@@ -0,0 +1,37 @@
+timestamp=20260606_112513
+repo_path=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/flow_grpo_cxr
+config_entry=config/grpo.py:general_radiomics_omnigen_4gpu_kl
+server_profile=
+omnigen_code_root=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint
+sft_lora_path=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint/results_new/scratch_15k
+dataset_root=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset
+train_jsonl=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_train.jsonl
+test_jsonl=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_test.jsonl
+dataset_path_remap_from=/home/wenting/dataset/LIDC_IDRI
+dataset_path_remap_to=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset
+mirp_repo_root=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/mirp
+save_dir=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/flow_grpo_cxr/logs/radiomics/img_only_manual/statcm_n005_bs64_steps200
+output_dir=outputs/img_only_manual/statcm_n005_bs64_steps200
+num_steps=10
+eval_num_steps=50
+noise_level=0.005
+sde_type=cps
+guidance_scale=2.5
+img_guidance_scale=2.0
+num_image_per_prompt=4
+train_batch_size=64
+test_batch_size=32
+num_batches_per_epoch=2
+kl_beta=0.005
+learning_rate=1e-4
+max_train_steps=0
+reward_feature_prefixes=cm_,stat_
+reward_base_feature_families=cm,stat
+reward_mask_channels=0,1
+num_processes=4
+wandb_project=flow_grpo_img_only_scratch30k
+wandb_group=manual_img_only
+wandb_name=statcm_n005_bs64_steps200
+wandb_tags=img-only,scratch30k,manual,noise0.005,bs64
+wandb_mode=online
+wandb_resume=never
diff --git a/radiomics/img_only_manual/statcm_n005_bs64_steps200/resolved_config.json b/radiomics/img_only_manual/statcm_n005_bs64_steps200/resolved_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..f3a3d85905deaa83269c973eb019ad382f3c27af
--- /dev/null
+++ b/radiomics/img_only_manual/statcm_n005_bs64_steps200/resolved_config.json
@@ -0,0 +1,143 @@
+{
+  "activation_checkpointing": true,
+  "allow_tf32": true,
+  "config_entry": "config/grpo.py:general_radiomics_omnigen_4gpu_kl",
+  "dataset": {
+    "dataset_path_remap_from": "/home/wenting/dataset/LIDC_IDRI",
+    "dataset_path_remap_to": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset",
+    "path_replacements": [
+      [
+        "/home/wenting/dataset/LIDC_IDRI",
+        "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset"
+      ]
+    ],
+    "root": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset",
+    "test_jsonl": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_test.jsonl",
+    "train_jsonl": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_train.jsonl"
+  },
+  "debug": false,
+  "eval_freq": 30,
+  "fsdp_optimizer_offload": false,
+  "logdir": "logs",
+  "mixed_precision": "bf16",
+  "num_checkpoint_limit": 5,
+  "num_epochs": 100000,
+  "per_prompt_stat_tracking": true,
+  "pretrained": {
+    "local_code_root": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint",
+    "model": "Shitao/OmniGen-v1",
+    "revision": "main",
+    "vae_path": null
+  },
+  "prompt_fn": "radiomics_edit",
+  "prompt_fn_kwargs": {},
+  "resolution": 256,
+  "resume_from": null,
+  "reward_fn": {
+    "radiomics": {
+      "base_feature_families": [
+        "cm",
+        "stat"
+      ],
+      "data_root": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset",
+      "dataset_path_remap_from": "/home/wenting/dataset/LIDC_IDRI",
+      "dataset_path_remap_to": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset",
+      "distance_metric": "relative_l1",
+      "feature_prefixes": [
+        "cm_",
+        "stat_"
+      ],
+      "fixed_bin_number": 32,
+      "gt_root": null,
+      "mask_channel": 0,
+      "mask_channels": [
+        0,
+        1
+      ],
+      "mask_root": null,
+      "mirp_repo_root": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/mirp",
+      "missing_score": -10.0,
+      "pixel_spacing_mm": 1.0,
+      "relative_error_clip": 10.0,
+      "reward_scale": 1.0,
+      "roi_name": "lung_roi",
+      "weight": 1.0
+    }
+  },
+  "run_name": "2026.06.06_11.27.47",
+  "sample": {
+    "batch_size": 8,
+    "eval_guidance_scale": 2.5,
+    "eval_img_guidance_scale": 2.0,
+    "eval_num_steps": 50,
+    "global_std": true,
+    "guidance_scale": 2.5,
+    "img_guidance_scale": 2.0,
+    "max_eval_batches": 2,
+    "max_input_image_size": 256,
+    "noise_level": 0.005,
+    "num_batches_per_epoch": 2,
+    "num_image_per_prompt": 4,
+    "num_steps": 10,
+    "same_latent": false,
+    "sde_type": "cps",
+    "sde_window_range": [
+      0,
+      10
+    ],
+    "sde_window_size": 2,
+    "test_batch_size": 32,
+    "train_batch_size": 64,
+    "use_img_guidance": true,
+    "use_input_image_size_as_output": false
+  },
+  "save_dir": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/flow_grpo_cxr/logs/radiomics/img_only_manual/statcm_n005_bs64_steps200",
+  "save_freq": 5,
+  "seed": 42,
+  "stage2_extra": {
+    "base_model": "Shitao/OmniGen-v1",
+    "mask_modules_path": null,
+    "omnigen_code_root": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint",
+    "reward_mask_channels": "0,1",
+    "rl_lora_target_modules": [
+      "qkv_proj",
+      "o_proj"
+    ],
+    "save_dir": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/flow_grpo_cxr/logs/radiomics/img_only_manual/statcm_n005_bs64_steps200",
+    "sft_lora_path": null,
+    "test_jsonl": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_test.jsonl",
+    "train_jsonl": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/dataset/anno/cxr_synth_anno_mask_train.jsonl"
+  },
+  "train": {
+    "adam_beta1": 0.9,
+    "adam_beta2": 0.999,
+    "adam_epsilon": 1e-08,
+    "adam_weight_decay": 0.0001,
+    "adv_clip_max": 5,
+    "batch_size": 64,
+    "beta": 0.005,
+    "cfg": true,
+    "clip_range": 0.0001,
+    "clip_range_gt": 0.0001,
+    "clip_range_lt": 0.0001,
+    "condition_dropout_prob": 0.01,
+    "ema": false,
+    "gradient_accumulation_steps": 1,
+    "learning_rate": 0.0001,
+    "lora_alpha": 64,
+    "lora_path": null,
+    "lora_rank": 32,
+    "max_grad_norm": 1.0,
+    "max_train_steps": 0,
+    "merge_lora_path": "/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint/results_new/scratch_15k",
+    "num_inner_epochs": 1,
+    "rl_lora_target_modules": [
+      "qkv_proj",
+      "o_proj"
+    ],
+    "timestep_fraction": 1.0,
+    "timestep_shift": 3.0,
+    "use_8bit_adam": false
+  },
+  "use_lora": true
+}
diff --git a/radiomics/img_only_sweeps/20260606_001259/imgonly_cm_n001_bs32_steps200/launch_config.txt b/radiomics/img_only_sweeps/20260606_001259/imgonly_cm_n001_bs32_steps200/launch_config.txt
new file mode 100644
index 0000000000000000000000000000000000000000..42909d44d4480ae4cb5a1c7a32602a679757cba7
--- /dev/null
+++ b/radiomics/img_only_sweeps/20260606_001259/imgonly_cm_n001_bs32_steps200/launch_config.txt
@@ -0,0 +1,30 @@
+timestamp=20260606_001259
+repo_path=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/flow_grpo_cxr
+config_entry=config/grpo.py:general_radiomics_omnigen_4gpu_kl
+omnigen_code_root=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint
+sft_lora_path=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint/results_new/pretrain_scratch/checkpoints/0030500
+save_dir=logs/radiomics/img_only_sweeps/20260606_001259/imgonly_cm_n001_bs32_steps200
+output_dir=outputs/img_only_sweeps/20260606_001259/imgonly_cm_n001_bs32_steps200
+num_steps=10
+eval_num_steps=50
+noise_level=0.01
+sde_type=cps
+guidance_scale=2.5
+img_guidance_scale=2.0
+num_image_per_prompt=4
+train_batch_size=32
+test_batch_size=16
+num_batches_per_epoch=2
+kl_beta=0.005
+learning_rate=1e-4
+max_train_steps=200
+reward_feature_prefixes=cm_
+reward_base_feature_families=cm
+reward_mask_channels=0,1
+num_processes=4
+wandb_project=flow_grpo_img_only_scratch15k
+wandb_group=img_only_pilot_20260606_001259
+wandb_name=imgonly_cm_n001_bs32_steps200-20260606_001259
+wandb_tags=img-only,scratch15k,cm,noisen001,bs32
+wandb_mode=online
+wandb_resume=never
diff --git a/radiomics/img_only_sweeps/20260606_002011/imgonly_cm_n001_bs32_steps200/launch_config.txt b/radiomics/img_only_sweeps/20260606_002011/imgonly_cm_n001_bs32_steps200/launch_config.txt
new file mode 100644
index 0000000000000000000000000000000000000000..c10fd833f7c5ea7cf5424783fbdada1103b4c9d9
--- /dev/null
+++ b/radiomics/img_only_sweeps/20260606_002011/imgonly_cm_n001_bs32_steps200/launch_config.txt
@@ -0,0 +1,30 @@
+timestamp=20260606_002011
+repo_path=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/flow_grpo_cxr
+config_entry=config/grpo.py:general_radiomics_omnigen_4gpu_kl
+omnigen_code_root=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint
+sft_lora_path=/apdcephfs_zwfy2/share_303944931/connorxian/Code/medUni/wt/zr/gen_joint/results_new/pretrain_scratch/checkpoints/0030500
+save_dir=logs/radiomics/img_only_sweeps/20260606_002011/imgonly_cm_n001_bs32_steps200
+output_dir=outputs/img_only_sweeps/20260606_002011/imgonly_cm_n001_bs32_steps200
+num_steps=10
+eval_num_steps=50
+noise_level=0.01
+sde_type=cps
+guidance_scale=2.5
+img_guidance_scale=2.0
+num_image_per_prompt=4
+train_batch_size=32
+test_batch_size=16
+num_batches_per_epoch=2
+kl_beta=0.005
+learning_rate=1e-4
+max_train_steps=200
+reward_feature_prefixes=cm_
+reward_base_feature_families=cm
+reward_mask_channels=0,1
+num_processes=4
+wandb_project=flow_grpo_img_only_scratch15k
+wandb_group=img_only_pilot_20260606_002011
+wandb_name=imgonly_cm_n001_bs32_steps200-20260606_002011
+wandb_tags=img-only,scratch15k,cm,noisen001,bs32
+wandb_mode=online
+wandb_resume=never