Duplicate from nvidia/GR00T-N1.7-3B

Co-authored-by: You Liang Tan <youliangt@users.noreply.huggingface.co>

.gitattributes

@@ -0,0 +1,35 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tar filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text

EXPLAINABILITY.md

@@ -0,0 +1,14 @@
+# **Explainability**
+
+|Field:|Response:|  
+|:---:|:---:|  
+|Intended Domain:| Open foundation model for generalized humanoid robot reasoning and skills.|  
+|Model Type: |Robot VLA model|  
+|Intended Users:|This model is intended for developers and community that build and finetune robot foundation models.|  
+|Output:|The model outputs are actions, and the units are floating points. This is referred to as "robot action policy." Actions consist of continuous-value vectors that correspond to different motor controls on a robot.|  
+|Describe how the model works:|Accepts vision, language and proprioception, outputs robot action policy.|  
+|Technical Limitations & Mitigation:| This model is not tested or intended for use in mission critical applications that require functional safety. The use of the model in those applications is at the user's own risk and sole responsibility, including taking the necessary steps to add needed guardrails or safety mechanisms prior to deployment.<br><br>Risk: Model underperformance in highly dynamic environments with varying robot surroundings (e.g. furniture, objects, etc) and lighting conditions.<br>Mitigation: Enhance dataset with dynamic obstacle scenarios and fine-tune models accordingly.<br><br>Risk: Integration challenges in specific customer environments with varying robot surroundings (e.g. furniture, objects, etc) and lighting conditions.<br>Mitigation: Provide detailed integration guides and support, leveraging NVIDIA's ecosystem.<br><br>Risk: Limited initial support for certain robot embodiments.<br>Mitigation: Expand testing and validation across a wider range of robot platforms.|  
+|Verified to have met prescribed quality standards?|Yes|  
+|Performance Metrics:|Success rate, as well as the following:<br>1) if the trajectory is smooth and does not jitter<br>2) if the robot does not hit any other objects<br>3) if the trajectory is natural|  
+|Potential Known Risks:|This model is not tested or intended for use in mission critical applications that require functional safety. The use of the model in those applications is at the user's own risk and sole responsibility, including taking the necessary steps to add needed guardrails or safety mechanisms prior to deployment.|  
+|End User License Agreement:| Your use of this model is governed by the [NSCL V1 License](https://developer.download.nvidia.com/licenses/NVIDIA-OneWay-Noncommercial-License-22Mar2022.pdf?t=eyJscyI6ImdzZW8iLCJsc2QiOiJodHRwczovL3d3dy5nb29nbGUuY29tLyIsIm5jaWQiOiJzby15b3V0LTg3MTcwMS12dDQ4In0=).|

LICENSE

@@ -0,0 +1,61 @@
+NVIDIA License
+1. Definitions
+“Licensor” means any person or entity that distributes its Work.
+“Work” means (a) the original work of authorship made available under this license, 
+which may include software, documentation, or other files, and (b) any additions to or 
+derivative works  thereof  that are made available under this license.
+The terms “reproduce,” “reproduction,” “derivative works,” and “distribution” have the 
+meaning as provided under U.S. copyright law; provided, however, that for the purposes
+of this license, derivative works shall not include works that remain separable from, or 
+merely link (or bind by name) to the interfaces of, the Work.
+Works are “made available” under this license by including in or with the Work either (a)
+a copyright notice referencing the applicability of this license to the Work, or (b) a copy 
+of this license.
+2. License Grant
+2.1 Copyright Grant. Subject to the terms and conditions of this license, each 
+Licensor grants to you a perpetual, worldwide, non-exclusive, royalty-free, 
+copyright license to use, reproduce, prepare derivative works of, publicly display, 
+publicly perform, sublicense and distribute its Work and any resulting derivative 
+works in any form.
+3. Limitations
+3.1 Redistribution. You may reproduce or distribute the Work only if (a) you do so 
+under this license, (b) you include a complete copy of this license with your 
+distribution, and (c) you retain without modification any copyright, patent, 
+trademark, or attribution notices that are present in the Work.
+3.2 Derivative Works. You may specify that additional or different terms apply to 
+the use, reproduction, and distribution of your derivative works of the Work (“Your
+Terms”) only if (a) Your Terms provide that the use limitation in Section 3.3 
+applies to your derivative works, and (b) you identify the specific derivative works 
+that are subject to Your Terms. Notwithstanding Your Terms, this license (including
+the redistribution requirements in Section 3.1) will continue to apply to the Work 
+itself.
+3.3 Use Limitation. The Work and any derivative works thereof only may be used 
+or intended for use non-commercially. Notwithstanding the foregoing, NVIDIA 
+Corporation and its affiliates may use the Work and any derivative works 
+commercially. As used herein, “non-commercially” means for research or 
+evaluation purposes only.
+3.4 Patent Claims. If you bring or threaten to bring a patent claim against any 
+Licensor (including any claim, cross-claim or counterclaim in a lawsuit) to enforce 
+any patents that you allege are infringed by any Work, then your rights under this 
+license from such Licensor (including the grant in Section 2.1) will terminate 
+immediately.
+3.5 Trademarks. This license does not grant any rights to use any Licensor’s or its 
+affiliates’ names, logos, or trademarks, except as necessary to reproduce the 
+notices described in this license.
+3.6 Termination. If you violate any term of this license, then your rights under this 
+license (including the grant in Section 2.1) will terminate immediately.
+4. Disclaimer of Warranty.
+THE WORK IS PROVIDED “AS IS” WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, 
+EITHER EXPRESS OR IMPLIED, INCLUDING WARRANTIES OR CONDITIONS OF
+MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE, TITLE OR NON-
+INFRINGEMENT. YOU BEAR THE RISK OF UNDERTAKING ANY ACTIVITIES UNDER THIS 
+LICENSE.
+5. Limitation of Liability.
+EXCEPT AS PROHIBITED BY APPLICABLE LAW, IN NO EVENT AND UNDER NO LEGAL 
+THEORY, WHETHER IN TORT (INCLUDING NEGLIGENCE), CONTRACT, OR OTHERWISE 
+SHALL ANY LICENSOR BE LIABLE TO YOU FOR DAMAGES, INCLUDING ANY DIRECT, 
+INDIRECT, SPECIAL, INCIDENTAL, OR CONSEQUENTIAL DAMAGES ARISING OUT OF OR 
+RELATED TO THIS LICENSE, THE USE OR INABILITY TO USE THE WORK (INCLUDING BUT 
+NOT LIMITED TO LOSS OF GOODWILL, BUSINESS INTERRUPTION, LOST PROFITS OR 
+DATA, COMPUTER FAILURE OR MALFUNCTION, OR ANY OTHER DAMAGES OR LOSSES), 
+EVEN IF THE LICENSOR HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES.

PRIVACY.md

@@ -0,0 +1,11 @@
+# **Privacy**
+
+|Field:|Response:|  
+|:---:|:---:|  
+|Generatable or reverse engineerable personal data?|None|   
+|Personal data used to create this model?|No|  
+|How often is dataset reviewed?|Before Release|  
+|Is there provenance for all datasets used in training?|Yes|  
+|Does data labeling (annotation, metadata) comply with privacy laws?|Yes|
+|Is data compliant with data subject requests for data correction or removal, if such a request was made?|Yes|
+|Applicable NVIDIA Privacy Policy|https://www.nvidia.com/en-us/about-nvidia/privacy-policy/|

README.md

@@ -0,0 +1,232 @@
+---
+tags:
+- robotics
+---
+
+<div align="center">
+  <a href="https://github.com/NVIDIA/Isaac-GR00T">
+    <img src="https://cdn-uploads.huggingface.co/production/uploads/67b8da81d01134f89899b4a7/8bFQa2ZIGCsOQQ2ho2N_U.png">
+  </a>
+  <div align="center">
+  <a href="https://github.com/NVIDIA/Isaac-GR00T">
+    <img src="https://img.shields.io/badge/GitHub-grey?logo=GitHub" alt="GitHub Badge">
+  </a>
+  <a href="https://developer.nvidia.com/isaac/gr00t">
+    <img src="https://img.shields.io/badge/Website-green" alt="Website Badge">
+  </a>
+  <!-- <a href=""">
+    <img src="https://img.shields.io/badge/Project%20Page-blue?style=plastic" alt="Project Page Badge">
+  </a>
+  <a href="">
+    <img src="https://img.shields.io/badge/Research_Blog-black?style=flat" alt="Research Blog Badge">
+  </a>
+  <a href="">
+    <img src="https://img.shields.io/badge/Dataset-Overview-brightgreen?logo=googleforms" alt="Research Blog Badge">
+  </a>
+  -->
+</div>
+</div>
+
+# Model Overview
+
+<p align="center">
+  <img src="https://cdn-uploads.huggingface.co/production/uploads/67b8da81d01134f89899b4a7/ZCLLXZk2LQBG0YH_BmiIN.gif"
+       style="width:100%; max-width:1000px; height:auto;">
+</p>
+
+## Description:
+NVIDIA Isaac GR00T N1.7 is an open foundation model for generalized humanoid robot reasoning and skills. This cross-embodiment model takes multimodal input, including language and images, to perform manipulation tasks in diverse environments. Developers and researchers can post-train GR00T N1.7 with real or synthetic data for their specific humanoid robot or task.
+
+Isaac GR00T N1.7 is the medium-sized version of our model built using pre-trained vision and language encoders, and uses a flow matching action transformer to model a chunk of actions conditioned on vision, language and proprioception. 
+
+A detailed description of the Isaac GR00T N1.X architecture is provided in the GROOT N1 White Paper (https://arxiv.org/abs/2503.14734). 
+
+This model is ready for commercial/non-commercial use.
+
+**Model Developer**: NVIDIA
+
+## Model Versions
+The Isaac GR00T N1.7 model family includes the following 4 models:
+
+### GR00T N1.7 – SimplerEnv Bridge
+
+**Description**  
+N1.7 post-trained model using the **Bridge Dataset** in SimplerEnv.
+
+**Post-Training Data**  
+https://huggingface.co/datasets/IPEC-COMMUNITY/bridge_orig_lerobot
+
+**Dataset Summary**  
+A LeRobot-format conversion of **BridgeData V2**, originally containing **60,096 trajectories** of robot manipulation across **24 environments**.
+
+### GR00T N1.7 – SimplerEnv Fractal
+
+**Description**  
+N1.7 post-trained model using the **Fractal Dataset** in SimplerEnv.
+
+**Post-Training Data**  
+https://huggingface.co/datasets/IPEC-COMMUNITY/bridge_orig_lerobot
+
+**Dataset Summary**  
+A LeRobot-format conversion of **BridgeData V2**, originally containing **60,096 trajectories** of robot manipulation across **24 environments**.
+
+### GR00T N1.7 – Droid
+**Description**  
+N1.7 post-trained model using the **DROID Dataset**.
+
+**Post-Training Data**  
+https://droid-dataset.github.io/
+
+**Dataset Summary**  
+A large-scale **“in-the-wild” robot manipulation dataset** with approximately **76,000 demonstration trajectories (~350 hours)** of interaction data, collected across **564 distinct scenes in 52 buildings**, covering **86 manipulation tasks** from natural-language instructions.
+
+### GR00T N1.7 – LIBERO
+**Description**  
+N1.7 post-trained model using the **LIBERO Dataset**.
+
+**Post-Training Data**  
+https://github.com/Lifelong-Robot-Learning/LIBERO
+
+**Dataset Summary**  
+A benchmark for **lifelong robot learning**, providing **130 language-conditioned manipulation tasks** grouped into multiple task suites.  
+Includes **human-teleoperated demonstrations** designed to evaluate **knowledge transfer and continual learning** in robotic agents.
+
+## License
+This model is released under the  [NVIDIA Open Model License Agreement](https://www.nvidia.com/en-us/agreements/enterprise-software/nvidia-open-model-license/).
+
+
+### Deployment Geography:
+Global
+
+### Use Case:
+Researchers, Academics, Open-Source Community: AI-driven robotics research and algorithm development.
+Developers: Integrate and customize AI for various robotic applications.
+Startups & Companies: Accelerate robotics development and reduce training costs.
+
+### Release Date:
+* Github via https://github.com/NVIDIA/Isaac-GR00T
+* Huggingface via https://huggingface.co/collections/nvidia/gr00t-n17 
+
+## Computational Load (Internal Only: For NVIDIA Models Only)
+Cumulative Compute: Follow Instructions
+ Estimated Energy and Emissions for Model Training: Follow Instructions
+Total kWh:
+64 GB200 nodes * 4 gpus per node x 1200W x 0.001 x 0.8 x 120 hours * 1.4 =  41288 kWh
+Total Emission:
+410.5 * 41288 * 0.000001 = 16.949 tCO2e
+
+## Model Architecture:
+
+**GR00T-N1.7 VLM backbone is now [Cosmos-Reason2-2B](https://huggingface.co/nvidia/Cosmos-Reason2-8B)**
+
+**Network Architecture:** 
+
+The schematic diagram is shown in the illustration above.
+Red, Green, Blue (RGB) camera frames are processed through a pre-trained vision transformer (SigLip2).
+Text is encoded by a pre-trained transformer (T5)
+Robot proprioception is encoded using a multi-layer perceptron (MLP) indexed by the embodiment ID. To handle variable-dimension proprioception, inputs are padded to a configurable max length before feeding into the MLP.
+Actions are encoded and velocity predictions decoded by an MLP, one per unique embodiment.
+The flow matching transformer is implemented as a diffusion transformer (DiT), in which the diffusion step conditioning is implemented using adaptive layernorm (AdaLN).
+
+![Model Architecture](model-architecture.png)
+
+**Number of Model Parameters:** 3,000,000,000
+
+## Input:
+**Input Type(s):**
+-Vision: Image Frames
+-State: Robot Proprioception
+-Language Instruction: Text
+-Embodiment ID: Integer
+
+**Input Format:**
+-Vision: Variable number of uint8 image frames, coming from robot cameras
+-State: Floating Point
+-Language Instruction: String
+-Embodiment ID: Integer indicating which of the training embodiments is observed
+
+**Input Parameters:**
+-Vision: Two-Dimensional (2D) - Red, Green, Blue (RGB)
+-State: One-Dimensional (1D) - Floating number vector
+-Language Instruction: One-Dimensional (1D) - String
+-Embodiment ID: One-Dimensional (1D) - Integer
+
+## Output:
+**Output Type(s):** Actions
+
+**Output Format** Continuous-value vectors
+
+**Output Parameters:** [Two-Dimensional (2D)] <br>
+
+**Other Properties Related to Output:** Continuous-value vectors correspond to different motor controls on a robot, which depends on Degrees of Freedom of the robot embodiment.
+
+Our AI models are designed and/or optimized to run on NVIDIA GPU-accelerated systems. By leveraging NVIDIA’s hardware (e.g. GPU cores) and software frameworks (e.g., CUDA libraries), the model achieves faster training and inference times compared to CPU-only solutions. <br>
+
+## Software Integration:
+
+**Runtime Engine(s):** PyTorch
+
+**Supported Hardware Microarchitecture Compatibility:** 
+All of the below:
+* NVIDIA Ampere
+* NVIDIA Blackwell
+* NVIDIA Jetson
+* NVIDIA Hopper
+* NVIDIA Lovelace
+
+**[Preferred/Supported] Operating System(s):**
+* Linux
+
+The integration of foundation and fine-tuned models into AI systems requires additional testing using use-case-specific data to ensure safe and effective deployment. Following the V-model methodology, iterative testing and validation at both unit and system levels are essential to mitigate risks, meet technical and functional requirements, and ensure compliance with safety and ethical standards before deployment.
+
+# Model Version
+GR00T N1.7 EA
+
+# Training and Evaluation Datasets:
+The total size (in number of data points): 21.6 million <br>
+Total number of datasets: 13 <br>   
+
+
+## Training Dataset:
+GR00T Pretraining Data
+
+**Data Collection Method by dataset:** Hybrid: Human, Robot, Simulated.
+
+**Labeling Method by dataset:** Hybrid: Human, Automated.
+
+**Properties:** 
+* Cross-embodiment: Data collected on various robot embodiments 
+* Sensor types: RGB camera, robot proprioception, robot actuator data
+
+
+## Evaluation:
+We evaluate in both simulation and real robot benchmarks, as defined in the White Paper (https://arxiv.org/abs/2503.14734). 
+
+**Data Collection Method by dataset:** Hybrid: Human, Robot, Simulated.
+
+**Labeling Method by dataset:** Hybrid: Human, Automated.
+
+* Sim evaluation benchmarks for upper body control
+* 9 DexMG Whitepaper tasks
+* 24 RoboCasa simulated mobile manipulator tasks
+* 24 Digital Cousin simulated GR-1 humanoid manipulation tasks 
+* For sim, we automatically measure the success rate in each manipulation behavior.
+* For real robot
+    * Grocery packing task
+    * Novel objects (unseen from training data)
+    * Industrial multi-robot coordination with handoffs
+    * Evaluated by human observers in the lab
+
+
+## Inference:
+**Engine:** PyTorch
+**Test Hardware:** A6000
+
+## Ethical Considerations:
+NVIDIA believes Trustworthy AI is a shared responsibility and we have established policies and practices to enable development for a wide array of AI applications.  When downloaded or used in accordance with our terms of service, developers should work with their internal model team to ensure this model meets requirements for the relevant industry and use case and addresses unforeseen product misuse.  
+
+Users are responsible for model inputs and outputs. Users are responsible for ensuring safe integration of this model, including implementing guardrails as well as other safety mechanisms, prior to deployment.
+
+For more detailed information on ethical considerations for this model, please see the Model Card++ Explainability, Bias, Safety & Security, and Privacy Subcards.
+
+Please report model quality, risk, security vulnerabilities or NVIDIA AI Concerns [here](https://www.nvidia.com/en-us/support/submit-security-vulnerability/).

SAFETY_and_SECURITY.md

@@ -0,0 +1,8 @@
+# **Safety & Security**
+
+|Field:|Response:|  
+|:---:|:---:|  
+|Model Application(s):|Machinery and Robotics<br>Robot VLA - single-arm manipulation, bimanual grippers, bi-manual dex hands manipulation and humanoid dexterous manipulation|  
+|Describe life critical application (if present):|This model is not tested or intended for use in mission critical applications that require functional safety. The use of the model in those applications is at the user's own risk and sole responsibility, including taking the necessary steps to add needed guardrails or safety mechanisms prior to deployment.|  
+|Use Case Restrictions:|Abide by the [NSCL V1 License](https://developer.download.nvidia.com/licenses/NVIDIA-OneWay-Noncommercial-License-22Mar2022.pdf?t=eyJscyI6ImdzZW8iLCJsc2QiOiJodHRwczovL3d3dy5nb29nbGUuY29tLyIsIm5jaWQiOiJzby15b3V0LTg3MTcwMS12dDQ4In0=)|  
+|Model and Dataset Restrictions:|The Principle of least privilege (PoLP) is applied limiting access for dataset generation and model development. Restrictions enforce dataset access during training, and dataset license constraints adhered to.|

config.json

@@ -0,0 +1,86 @@
+{
+  "action_horizon": 40,
+  "add_pos_embed": true,
+  "apply_sincos_state_encoding": false,
+  "architectures": [
+    "Gr00tN1d7"
+  ],
+  "attn_dropout": 0.2,
+  "backbone_embedding_dim": 2048,
+  "color_jitter_params": {
+    "brightness": 0.3,
+    "contrast": 0.4,
+    "hue": 0.08,
+    "saturation": 0.5
+  },
+  "crop_fraction": 0.95,
+  "diffusion_model_cfg": {
+    "attention_head_dim": 48,
+    "dropout": 0.2,
+    "final_dropout": true,
+    "interleave_self_attention": true,
+    "norm_type": "ada_norm",
+    "num_attention_heads": 32,
+    "num_layers": 32,
+    "output_dim": 1024,
+    "positional_embeddings": null
+  },
+  "dtype": "bfloat16",
+  "exclude_state": false,
+  "formalize_language": true,
+  "hidden_size": 1024,
+  "image_crop_size": [
+    230,
+    230
+  ],
+  "image_target_size": [
+    256,
+    256
+  ],
+  "letter_box_transform": false,
+  "load_bf16": true,
+  "max_action_dim": 132,
+  "max_num_embodiments": 32,
+  "max_seq_len": 1024,
+  "max_state_dim": 132,
+  "model_dtype": "bfloat16",
+  "model_type": "Gr00tN1d7",
+  "noise_beta_alpha": 1.5,
+  "noise_beta_beta": 1.0,
+  "noise_s": 0.999,
+  "num_inference_timesteps": 4,
+  "num_timestep_buckets": 1000,
+  "random_history_crop": true,
+  "random_rotation_angle": 0,
+  "reproject_vision": false,
+  "rtc_ramp_rate": 6.0,
+  "select_layer": 16,
+  "shortest_image_edge": 256,
+  "state_dropout_prob": 0.2,
+  "state_gaussian_noise_std": 0.0,
+  "transformers_version": "4.57.1",
+  "tune_diffusion_model": true,
+  "tune_linear": true,
+  "tune_llm": true,
+  "tune_projector": true,
+  "tune_top_llm_layers": 0,
+  "tune_visual": true,
+  "tune_vlln": true,
+  "use_albumentations": true,
+  "use_alternate_vl_dit": true,
+  "use_flash_attention": true,
+  "use_future_tokens": false,
+  "use_mean_std": false,
+  "use_percentiles": true,
+  "use_vl_self_attention": true,
+  "use_vlln": true,
+  "vl_self_attention_cfg": {
+    "attention_head_dim": 64,
+    "dropout": 0.2,
+    "final_dropout": true,
+    "num_attention_heads": 32,
+    "num_layers": 4,
+    "positional_embeddings": null
+  },
+  "model_name": "nvidia/Cosmos-Reason2-2B"
+}

embodiment_id.json

@@ -0,0 +1,54 @@
+{
+  "robocasa_panda_omron": 13,
+  "oxe_droid": 17,
+  "oxe_fractal": 18,
+  "oxe_language_table": 19,
+  "oxe_bridge": 20,
+  "unknown": 22,
+  "gr1_unified": 20,
+  "agibot": 26,
+  "sim_behavior_r1_pro": 23,
+  "xdof": 24,
+  "xdof_oss_data": 25,
+  "unitree_g1_full_body_with_waist_height_nav_cmd": 25,
+  "real_r1_pro_sharpa": 27,
+  "real_r1_pro_sharpa_add_view": 27,
+  "real_r1_pro_sharpa_relative_arm_joint": 26,
+  "real_r1_pro_sharpa_delta_eef": 26,
+  "real_r1_pro_sharpa_absolute_eef": 26,
+  "real_r1_pro_sharpa_meanstd": 26,
+  "real_r1_pro_sharpa_relative_eef": 26,
+  "real_r1_pro_sharpa_relative_eef_add_view": 26,
+  "real_r1_pro_sharpa_relative_eef_relative_hand": 26,
+  "real_r1_pro_sharpa_relative_eef_human": 26,
+  "real_r1_pro_sharpa_relative_eef_human_add_view": 26,
+  "real_r1_pro_sharpa_relative_eef_human_relative_hand": 26,
+  "real_r1_pro_sharpa_relative_eef_egodex": 26,
+  "real_r1_pro_sharpa_relative_eef_egodex_relative_hand": 26,
+  "real_r1_pro_sharpa_relative_eef_egodex_wrist_only": 26,
+  "real_r1_pro_sharpa_relative_eef_maxinsights": 26,
+  "real_r1_pro_sharpa_relative_eef_maxinsights_relative_hand": 26,
+  "real_r1_pro_sharpa_relative_eef_mecka": 26,
+  "real_r1_pro_sharpa_relative_eef_mecka_relative_hand": 26,
+  "real_g1_relative_eef_absolute_joints": 25,
+  "real_g1_relative_eef_absolute_joints_wrist_cam": 25,
+  "real_g1_relative_eef_relative_joints": 25,
+  "real_r1_pro_sharpa_relative_eef_relative_hand_relative_joint": 26,
+  "real_r1_pro_sharpa_relative_joint": 29,
+  "oxe_droid_relative_eef_relative_joint": 24,
+  "oxe_droid_relative_eef_relative_joint_swapped": 24,
+  "oxe_droid_relative_eef_relative_joint_upweight_z": 24,
+  "oxe_droid_relative_eef_relative_joint_upweight_z_swapped": 24,
+  "oxe_droid_relative_eef_relative_joint_3view": 24,
+  "oxe_droid_relative_eef_relative_joint_3view_swapped": 24,
+  "oxe_droid_relative_eef": 24,
+  "oxe_droid_joint_position_relative": 24,
+  "xdof_relative_eef_relative_joint": 27,
+  "xdof_relative_eef_relative_joint_subtask": 27,
+  "xdof_relative_eef": 27,
+  "xdof_relative_joint": 28,
+  "simpler_env_google": 0,
+  "simpler_env_widowx": 1,
+  "libero_sim": 2,
+  "droid_sim": 3
+}

experiment_cfg/conf.yaml

@@ -0,0 +1,1324 @@
+load_config_path: groot/vla/omni/configs/experiments/r1_pro/sharpa/n17_pretrain/n17_pretrain_human_robot_cross_embodiment_fix_yam_absolute_hand_2step.yaml
+model:
+  return_dict: true
+  output_hidden_states: false
+  torchscript: false
+  dtype: null
+  pruned_heads: {}
+  tie_word_embeddings: true
+  chunk_size_feed_forward: 0
+  is_encoder_decoder: false
+  is_decoder: false
+  cross_attention_hidden_size: null
+  add_cross_attention: false
+  tie_encoder_decoder: false
+  architectures: null
+  finetuning_task: null
+  id2label:
+    0: LABEL_0
+    1: LABEL_1
+  label2id:
+    LABEL_0: 0
+    LABEL_1: 1
+  task_specific_params: null
+  problem_type: null
+  tokenizer_class: null
+  prefix: null
+  bos_token_id: null
+  pad_token_id: null
+  eos_token_id: null
+  sep_token_id: null
+  decoder_start_token_id: null
+  max_length: 20
+  min_length: 0
+  do_sample: false
+  early_stopping: false
+  num_beams: 1
+  temperature: 1.0
+  top_k: 50
+  top_p: 1.0
+  typical_p: 1.0
+  repetition_penalty: 1.0
+  length_penalty: 1.0
+  no_repeat_ngram_size: 0
+  encoder_no_repeat_ngram_size: 0
+  bad_words_ids: null
+  num_return_sequences: 1
+  output_scores: false
+  return_dict_in_generate: false
+  forced_bos_token_id: null
+  forced_eos_token_id: null
+  remove_invalid_values: false
+  exponential_decay_length_penalty: null
+  suppress_tokens: null
+  begin_suppress_tokens: null
+  num_beam_groups: 1
+  diversity_penalty: 0.0
+  transformers_version: null
+  model_type: GrootN1d5Qwen
+  model_dtype: bfloat16
+  vlm_backend: qwen3
+  vlm_model_path: nvidia/Cosmos-Reason2-2B
+  backbone_embedding_dim: 2048
+  tune_llm: false
+  tune_top_llm_layers: 0
+  tune_visual: false
+  tune_linear: true
+  select_layer: 16
+  reproject_vision: false
+  use_flash_attention: true
+  load_bf16: true
+  exclude_state: false
+  image_crop_size:
+  - 230
+  - 230
+  image_target_size:
+  - 256
+  - 256
+  random_rotation_angle: 0
+  color_jitter_params:
+    brightness: 0.3
+    contrast: 0.4
+    saturation: 0.5
+    hue: 0.08
+  formalize_language: true
+  action_space_prompt: false
+  apply_sincos_state_encoding: false
+  letter_box_transform: false
+  use_percentiles: true
+  use_mean_std: false
+  use_albumentations: true
+  shortest_image_edge: 256
+  crop_fraction: 0.95
+  random_history_crop: true
+  state_gaussian_noise_std: 0.0
+  do_human_interpolation: false
+  interpolation_steps: 20
+  human_embodiment_tags: null
+  max_state_dim: 132
+  max_action_dim: 132
+  action_horizon: 40
+  hidden_size: 1024
+  dit_latent_dim: 1536
+  state_dropout_prob: 0.2
+  language_dropout_prob: 0.0
+  add_pos_embed: true
+  attn_dropout: 0.2
+  use_vlln: true
+  use_vl_self_attention: true
+  max_seq_len: 1024
+  use_future_tokens: false
+  use_alternate_vl_dit: true
+  vl_self_attention_cfg:
+    positional_embeddings: null
+    num_layers: 4
+    num_attention_heads: 32
+    attention_head_dim: 64
+    dropout: 0.2
+    final_dropout: true
+  diffusion_model_cfg:
+    positional_embeddings: null
+    num_layers: 32
+    num_attention_heads: 32
+    attention_head_dim: 48
+    norm_type: ada_norm
+    dropout: 0.2
+    final_dropout: true
+    output_dim: 1024
+    interleave_self_attention: true
+    cross_attention_dim: 2048
+  num_inference_timesteps: 4
+  noise_beta_alpha: 1.5
+  noise_beta_beta: 1.0
+  noise_s: 0.999
+  num_timestep_buckets: 1000
+  tune_projector: true
+  tune_diffusion_model: true
+  tune_vlln: true
+  max_num_embodiments: 32
+  rtc_ramp_rate: 6.0
+  tf_legacy_loss: false
+  use_bfloat16: false
+data:
+  datasets:
+  - dataset_paths:
+    - /mnt/aws-lfs-02/shared/datasets/xdof.yam_v7_all_merged_global_task_exclude_bad_subtasks
+    embodiment_tag: xdof_relative_eef_relative_joint
+    mix_ratio: 0.1
+    dataset_type: physical_embodiment
+  - dataset_paths:
+    - /mnt/aws-lfs-02/shared/datasets/xdof.yam_v7_subtask_only_merged_global_task
+    embodiment_tag: xdof_relative_eef_relative_joint_subtask
+    mix_ratio: 0.2
+    dataset_type: physical_embodiment
+  - dataset_paths:
+    - /mnt/aws-lfs-02/shared/datasets/droid_101_success_idlefiltered_n17
+    - /mnt/aws-lfs-02/shared/datasets/droid_101_success_idlefiltered_n17_swapped
+    embodiment_tag: oxe_droid_relative_eef_relative_joint
+    mix_ratio: 0.1
+    dataset_type: physical_embodiment
+  - dataset_paths:
+    - /mnt/aws-lfs-02/shared/datasets/galaxea_sharpa/real_g1.g1-in-the-wild-merged
+    embodiment_tag: real_g1_relative_eef_relative_joints
+    mix_ratio: 0.05
+    dataset_type: physical_embodiment
+  - dataset_paths:
+    - /mnt/aws-lfs-02/shared/datasets/galaxea_sharpa/real_r1_pro_sharpa.inlab_play_real_robot_batch_1
+    - /mnt/aws-lfs-02/shared/datasets/galaxea_sharpa/real_r1_pro_sharpa.inlab_play_real_robot_batch_2
+    - /mnt/aws-lfs-02/shared/datasets/galaxea_sharpa/real_r1_pro_sharpa.miscellaneous_1k_trajectories
+    embodiment_tag: real_r1_pro_sharpa_relative_eef
+    mix_ratio: 0.05
+    dataset_type: physical_embodiment
+  - dataset_paths:
+    - /mnt/aws-lfs-02/shared/datasets/galaxea_sharpa/mecka_lerobot/real_r1_pro_sharpa.mecka_batch1-2025-12-10-merged
+    - /mnt/aws-lfs-02/shared/datasets/galaxea_sharpa/mecka_lerobot/real_r1_pro_sharpa.mecka_batch3_2026-01-04-merged_backup
+    - /mnt/aws-lfs-02/shared/datasets/galaxea_sharpa/mecka_lerobot/real_r1_pro_sharpa.mecka_batch4_2026-01-05-merged_backup
+    - /mnt/aws-lfs-02/shared/datasets/galaxea_sharpa/mecka_lerobot/real_r1_pro_sharpa.mecka_batch5_2026-01-05-merged_backup
+    - /mnt/aws-lfs-02/shared/datasets/galaxea_sharpa/mecka_lerobot/real_r1_pro_sharpa.mecka_batch6_2026-01-05-merged_backup
+    - /mnt/aws-lfs-02/shared/datasets/galaxea_sharpa/mecka_lerobot/real_r1_pro_sharpa.mecka_batch10_2026-01-10-merged_backup
+    - /mnt/aws-lfs-02/shared/datasets/galaxea_sharpa/mecka_lerobot/real_r1_pro_sharpa.mecka_batch11_2026-01-10-merged_backup
+    - /mnt/aws-lfs-02/shared/datasets/galaxea_sharpa/mecka_lerobot/real_r1_pro_sharpa.mecka_batch12_2026-01-10-merged_backup
+    - /mnt/aws-lfs-02/shared/datasets/galaxea_sharpa/mecka_lerobot/real_r1_pro_sharpa.mecka_batch8_2026-01-10-merged_backup
+    - /mnt/aws-lfs-02/shared/datasets/galaxea_sharpa/mecka_lerobot/real_r1_pro_sharpa.mecka_batch9_2026-01-10-merged_backup
+    embodiment_tag: real_r1_pro_sharpa_relative_eef_mecka
+    mix_ratio: 0.25
+    dataset_type: physical_embodiment
+  - dataset_paths:
+    - /mnt/aws-lfs-02/shared/datasets/galaxea_sharpa/maxinsights_lerobot_updated/1530hrs/real_r1_pro_sharpa.maxinsights_1530hrs_updated_train_set_merged
+    embodiment_tag: real_r1_pro_sharpa_relative_eef_maxinsights
+    mix_ratio: 0.2
+    dataset_type: physical_embodiment
+  - dataset_paths:
+    - /mnt/aws-lfs-02/shared/datasets/galaxea_sharpa/real_r1_pro_sharpa.inlab_play_human_batch1
+    - /mnt/aws-lfs-02/shared/datasets/galaxea_sharpa/real_r1_pro_sharpa.inlab_play_human_batch2
+    - /mnt/aws-lfs-02/shared/datasets/galaxea_sharpa/real_r1_pro_sharpa.shirt_rolling_task24_2000_human_video_filter_n6_keep1619_demo_stats
+    - /mnt/aws-lfs-02/shared/datasets/galaxea_sharpa/real_r1_pro_sharpa.shirt_rolling_task15_2000_human_video_filter_n6_keep572_demo_stats
+    - /mnt/aws-lfs-02/shared/datasets/galaxea_sharpa/real_r1_pro_sharpa.sort_cards_human_filter_n6_keep523_demo_stats_overwrite_left_side_stats
+    - /mnt/aws-lfs-02/shared/datasets/galaxea_sharpa/real_r1_pro_sharpa.tong_task38_2000_human_video_overwrite_left_side_stats
+    - /mnt/aws-lfs-02/shared/datasets/galaxea_sharpa/real_r1_pro_sharpa.syringe_task30i_2000_human_video_filtered
+    - /mnt/aws-lfs-02/shared/datasets/galaxea_sharpa/real_r1_pro_sharpa.unscrew_bottle_task43_2000_human_video_fixed-duration
+    - /mnt/aws-lfs-02/shared/datasets/galaxea_sharpa/real_r1_pro_sharpa.unscrew_Jim_bottle_task47_600_human_video
+    - /mnt/aws-lfs-02/shared/datasets/galaxea_sharpa/real_r1_pro_sharpa.fold_shirt_task30b_500_human_video_halfdone
+    - /mnt/aws-lfs-02/shared/datasets/galaxea_sharpa/real_r1_pro_sharpa.fold_towel_task30c_500_human_video_halfdone
+    - /mnt/aws-lfs-02/shared/datasets/galaxea_sharpa/real_r1_pro_sharpa.sort_cards_task32e_1000_human_video
+    embodiment_tag: real_r1_pro_sharpa_relative_eef_human
+    mix_ratio: 0.05
+    dataset_type: physical_embodiment
+  modality_configs:
+    real_g1_relative_eef_relative_joints:
+      video:
+        delta_indices:
+        - -20
+        - 0
+        modality_keys:
+        - ego_view
+        normalization_mode: null
+        action_representation: null
+        exclude_state: false
+        action_type: null
+        action_format: null
+        normalize_rotation: true
+        wrist_keys: null
+        hand_keys: null
+        extra_keys: null
+        loss_weights: null
+      state:
+        delta_indices:
+        - 0
+        modality_keys:
+        - left_wrist_eef_9d
+        - right_wrist_eef_9d
+        - left_hand
+        - right_hand
+        - left_arm
+        - right_arm
+        - waist
+        normalization_mode: null
+        action_representation: null
+        exclude_state: false
+        action_type: null
+        action_format: null
+        normalize_rotation: true
+        wrist_keys: null
+        hand_keys: null
+        extra_keys: null
+        loss_weights: null
+      action:
+        delta_indices:
+        - 0
+        - 1
+        - 2
+        - 3
+        - 4
+        - 5
+        - 6
+        - 7
+        - 8
+        - 9
+        - 10
+        - 11
+        - 12
+        - 13
+        - 14
+        - 15
+        - 16
+        - 17
+        - 18
+        - 19
+        - 20
+        - 21
+        - 22
+        - 23
+        - 24
+        - 25
+        - 26
+        - 27
+        - 28
+        - 29
+        - 30
+        - 31
+        - 32
+        - 33
+        - 34
+        - 35
+        - 36
+        - 37
+        - 38
+        - 39
+        modality_keys:
+        - left_wrist_eef_9d
+        - right_wrist_eef_9d
+        - left_hand
+        - right_hand
+        - left_arm
+        - right_arm
+        - waist
+        - base_height_command
+        - navigate_command
+        normalization_mode: null
+        action_representation:
+        - {}
+        - {}
+        - {}
+        - {}
+        - {}
+        - {}
+        - {}
+        - {}
+        - {}
+        exclude_state: false
+        action_type:
+        - {}
+        - {}
+        - {}
+        - {}
+        - {}
+        - {}
+        - {}
+        - {}
+        - {}
+        action_format:
+        - {}
+        - {}
+        - {}
+        - {}
+        - {}
+        - {}
+        - {}
+        - {}
+        - {}
+        normalize_rotation: true
+        wrist_keys:
+        - left_wrist_eef_9d
+        - right_wrist_eef_9d
+        hand_keys:
+        - left_hand
+        - right_hand
+        extra_keys:
+        - left_arm
+        - right_arm
+        - waist
+        - base_height_command
+        - navigate_command
+        loss_weights: null
+      language:
+        delta_indices:
+        - 0
+        modality_keys:
+        - annotation.human.task_description
+        normalization_mode: null
+        action_representation: null
+        exclude_state: false
+        action_type: null
+        action_format: null
+        normalize_rotation: true
+        wrist_keys: null
+        hand_keys: null
+        extra_keys: null
+        loss_weights: null
+    real_r1_pro_sharpa_relative_eef_mecka:
+      video:
+        delta_indices:
+        - -30
+        - 0
+        modality_keys:
+        - ego_view_cropratio_res320x240_freq30
+        normalization_mode: null
+        action_representation: null
+        exclude_state: false
+        action_type: null
+        action_format: null
+        normalize_rotation: true
+        wrist_keys: null
+        hand_keys: null
+        extra_keys: null
+        loss_weights: null
+      state:
+        delta_indices:
+        - 0
+        modality_keys:
+        - left_wrist_eef
+        - right_wrist_eef
+        - left_hand_joints
+        - right_hand_joints
+        normalization_mode: null
+        action_representation: null
+        exclude_state: true
+        action_type: null
+        action_format: null
+        normalize_rotation: true
+        wrist_keys: null
+        hand_keys: null
+        extra_keys: null
+        loss_weights: null
+      action:
+        delta_indices:
+        - 0
+        - 1
+        - 2
+        - 3
+        - 4
+        - 5
+        - 6
+        - 7
+        - 8
+        - 9
+        - 10
+        - 11
+        - 12
+        - 13
+        - 14
+        - 15
+        - 16
+        - 17
+        - 18
+        - 19
+        - 20
+        - 21
+        - 22
+        - 23
+        - 24
+        - 25
+        - 26
+        - 27
+        - 28
+        - 29
+        - 30
+        - 31
+        - 32
+        - 33
+        - 34
+        - 35
+        - 36
+        - 37
+        - 38
+        - 39
+        modality_keys:
+        - left_wrist_eef
+        - right_wrist_eef
+        - left_hand_joints
+        - right_hand_joints
+        normalization_mode: null
+        action_representation:
+        - {}
+        - {}
+        - {}
+        - {}
+        exclude_state: false
+        action_type:
+        - {}
+        - {}
+        - {}
+        - {}
+        action_format:
+        - {}
+        - {}
+        - {}
+        - {}
+        normalize_rotation: true
+        wrist_keys:
+        - left_wrist_eef
+        - right_wrist_eef
+        hand_keys:
+        - left_hand_joints
+        - right_hand_joints
+        extra_keys: []
+        loss_weights: null
+      language:
+        delta_indices:
+        - 0
+        modality_keys:
+        - annotation.human.coarse_action
+        normalization_mode: null
+        action_representation: null
+        exclude_state: false
+        action_type: null
+        action_format: null
+        normalize_rotation: true
+        wrist_keys: null
+        hand_keys: null
+        extra_keys: null
+        loss_weights: null
+    oxe_droid_relative_eef_relative_joint:
+      video:
+        delta_indices:
+        - -15
+        - 0
+        modality_keys:
+        - exterior_image_1_left
+        - wrist_image_left
+        normalization_mode: null
+        action_representation: null
+        exclude_state: false
+        action_type: null
+        action_format: null
+        normalize_rotation: true
+        wrist_keys: null
+        hand_keys: null
+        extra_keys: null
+        loss_weights: null
+      state:
+        delta_indices:
+        - 0
+        modality_keys:
+        - eef_9d
+        - gripper_position
+        - joint_position
+        normalization_mode: null
+        action_representation: null
+        exclude_state: false
+        action_type: null
+        action_format: null
+        normalize_rotation: true
+        wrist_keys: null
+        hand_keys: null
+        extra_keys: null
+        loss_weights: null
+      action:
+        delta_indices:
+        - 0
+        - 1
+        - 2
+        - 3
+        - 4
+        - 5
+        - 6
+        - 7
+        - 8
+        - 9
+        - 10
+        - 11
+        - 12
+        - 13
+        - 14
+        - 15
+        - 16
+        - 17
+        - 18
+        - 19
+        - 20
+        - 21
+        - 22
+        - 23
+        - 24
+        - 25
+        - 26
+        - 27
+        - 28
+        - 29
+        - 30
+        - 31
+        - 32
+        - 33
+        - 34
+        - 35
+        - 36
+        - 37
+        - 38
+        - 39
+        modality_keys:
+        - eef_9d
+        - gripper_position
+        - joint_position
+        normalization_mode: null
+        action_representation:
+        - {}
+        - {}
+        - {}
+        exclude_state: false
+        action_type:
+        - {}
+        - {}
+        - {}
+        action_format:
+        - {}
+        - {}
+        - {}
+        normalize_rotation: true
+        wrist_keys:
+        - eef_9d
+        hand_keys:
+        - gripper_position
+        extra_keys:
+        - joint_position
+        loss_weights: null
+      language:
+        delta_indices:
+        - 0
+        modality_keys:
+        - annotation.language.language_instruction
+        - annotation.language.language_instruction_2
+        - annotation.language.language_instruction_3
+        normalization_mode: null
+        action_representation: null
+        exclude_state: false
+        action_type: null
+        action_format: null
+        normalize_rotation: true
+        wrist_keys: null
+        hand_keys: null
+        extra_keys: null
+        loss_weights: null
+    real_r1_pro_sharpa_relative_eef_human:
+      video:
+        delta_indices:
+        - -20
+        - 0
+        modality_keys:
+        - ego_view_res320x240_freq20
+        - left_wrist_view_res320x240_freq20
+        - right_wrist_view_res320x240_freq20
+        normalization_mode: null
+        action_representation: null
+        exclude_state: false
+        action_type: null
+        action_format: null
+        normalize_rotation: true
+        wrist_keys: null
+        hand_keys: null
+        extra_keys: null
+        loss_weights: null
+      state:
+        delta_indices:
+        - 0
+        modality_keys:
+        - left_wrist_eef
+        - right_wrist_eef
+        - left_hand_joints
+        - right_hand_joints
+        normalization_mode: null
+        action_representation: null
+        exclude_state: true
+        action_type: null
+        action_format: null
+        normalize_rotation: true
+        wrist_keys: null
+        hand_keys: null
+        extra_keys: null
+        loss_weights: null
+      action:
+        delta_indices:
+        - 0
+        - 1
+        - 2
+        - 3
+        - 4
+        - 5
+        - 6
+        - 7
+        - 8
+        - 9
+        - 10
+        - 11
+        - 12
+        - 13
+        - 14
+        - 15
+        - 16
+        - 17
+        - 18
+        - 19
+        - 20
+        - 21
+        - 22
+        - 23
+        - 24
+        - 25
+        - 26
+        - 27
+        - 28
+        - 29
+        - 30
+        - 31
+        - 32
+        - 33
+        - 34
+        - 35
+        - 36
+        - 37
+        - 38
+        - 39
+        modality_keys:
+        - left_wrist_eef
+        - right_wrist_eef
+        - left_hand_joints
+        - right_hand_joints
+        normalization_mode: null
+        action_representation:
+        - {}
+        - {}
+        - {}
+        - {}
+        exclude_state: false
+        action_type:
+        - {}
+        - {}
+        - {}
+        - {}
+        action_format:
+        - {}
+        - {}
+        - {}
+        - {}
+        normalize_rotation: true
+        wrist_keys:
+        - left_wrist_eef
+        - right_wrist_eef
+        hand_keys:
+        - left_hand_joints
+        - right_hand_joints
+        extra_keys: []
+        loss_weights: null
+      language:
+        delta_indices:
+        - 0
+        modality_keys:
+        - annotation.human.coarse_action
+        normalization_mode: null
+        action_representation: null
+        exclude_state: false
+        action_type: null
+        action_format: null
+        normalize_rotation: true
+        wrist_keys: null
+        hand_keys: null
+        extra_keys: null
+        loss_weights: null
+    xdof_relative_eef_relative_joint:
+      video:
+        delta_indices:
+        - -30
+        - 0
+        modality_keys:
+        - top_camera-images-rgb_320_240
+        - left_camera-images-rgb_320_240
+        - right_camera-images-rgb_320_240
+        normalization_mode: null
+        action_representation: null
+        exclude_state: false
+        action_type: null
+        action_format: null
+        normalize_rotation: true
+        wrist_keys: null
+        hand_keys: null
+        extra_keys: null
+        loss_weights: null
+      state:
+        delta_indices:
+        - 0
+        modality_keys:
+        - left_wrist_eef
+        - right_wrist_eef
+        - left_gripper_pos
+        - right_gripper_pos
+        - left_joint_pos
+        - right_joint_pos
+        normalization_mode: null
+        action_representation: null
+        exclude_state: false
+        action_type: null
+        action_format: null
+        normalize_rotation: true
+        wrist_keys: null
+        hand_keys: null
+        extra_keys: null
+        loss_weights: null
+      action:
+        delta_indices:
+        - 0
+        - 1
+        - 2
+        - 3
+        - 4
+        - 5
+        - 6
+        - 7
+        - 8
+        - 9
+        - 10
+        - 11
+        - 12
+        - 13
+        - 14
+        - 15
+        - 16
+        - 17
+        - 18
+        - 19
+        - 20
+        - 21
+        - 22
+        - 23
+        - 24
+        - 25
+        - 26
+        - 27
+        - 28
+        - 29
+        - 30
+        - 31
+        - 32
+        - 33
+        - 34
+        - 35
+        - 36
+        - 37
+        - 38
+        - 39
+        modality_keys:
+        - left_wrist_eef
+        - right_wrist_eef
+        - left_gripper_pos
+        - right_gripper_pos
+        - left_joint_pos
+        - right_joint_pos
+        normalization_mode: null
+        action_representation:
+        - {}
+        - {}
+        - {}
+        - {}
+        - {}
+        - {}
+        exclude_state: false
+        action_type:
+        - {}
+        - {}
+        - {}
+        - {}
+        - {}
+        - {}
+        action_format:
+        - {}
+        - {}
+        - {}
+        - {}
+        - {}
+        - {}
+        normalize_rotation: true
+        wrist_keys:
+        - left_wrist_eef
+        - right_wrist_eef
+        hand_keys:
+        - left_gripper_pos
+        - right_gripper_pos
+        extra_keys:
+        - left_joint_pos
+        - right_joint_pos
+        loss_weights: null
+      language:
+        delta_indices:
+        - 0
+        modality_keys:
+        - annotation.task
+        normalization_mode: null
+        action_representation: null
+        exclude_state: false
+        action_type: null
+        action_format: null
+        normalize_rotation: true
+        wrist_keys: null
+        hand_keys: null
+        extra_keys: null
+        loss_weights: null
+    xdof_relative_eef_relative_joint_subtask:
+      video:
+        delta_indices:
+        - -30
+        - 0
+        modality_keys:
+        - top_camera-images-rgb_320_240
+        - left_camera-images-rgb_320_240
+        - right_camera-images-rgb_320_240
+        normalization_mode: null
+        action_representation: null
+        exclude_state: false
+        action_type: null
+        action_format: null
+        normalize_rotation: true
+        wrist_keys: null
+        hand_keys: null
+        extra_keys: null
+        loss_weights: null
+      state:
+        delta_indices:
+        - 0
+        modality_keys:
+        - left_wrist_eef
+        - right_wrist_eef
+        - left_gripper_pos
+        - right_gripper_pos
+        - left_joint_pos
+        - right_joint_pos
+        normalization_mode: null
+        action_representation: null
+        exclude_state: false
+        action_type: null
+        action_format: null
+        normalize_rotation: true
+        wrist_keys: null
+        hand_keys: null
+        extra_keys: null
+        loss_weights: null
+      action:
+        delta_indices:
+        - 0
+        - 1
+        - 2
+        - 3
+        - 4
+        - 5
+        - 6
+        - 7
+        - 8
+        - 9
+        - 10
+        - 11
+        - 12
+        - 13
+        - 14
+        - 15
+        - 16
+        - 17
+        - 18
+        - 19
+        - 20
+        - 21
+        - 22
+        - 23
+        - 24
+        - 25
+        - 26
+        - 27
+        - 28
+        - 29
+        - 30
+        - 31
+        - 32
+        - 33
+        - 34
+        - 35
+        - 36
+        - 37
+        - 38
+        - 39
+        modality_keys:
+        - left_wrist_eef
+        - right_wrist_eef
+        - left_gripper_pos
+        - right_gripper_pos
+        - left_joint_pos
+        - right_joint_pos
+        normalization_mode: null
+        action_representation:
+        - {}
+        - {}
+        - {}
+        - {}
+        - {}
+        - {}
+        exclude_state: false
+        action_type:
+        - {}
+        - {}
+        - {}
+        - {}
+        - {}
+        - {}
+        action_format:
+        - {}
+        - {}
+        - {}
+        - {}
+        - {}
+        - {}
+        normalize_rotation: true
+        wrist_keys:
+        - left_wrist_eef
+        - right_wrist_eef
+        hand_keys:
+        - left_gripper_pos
+        - right_gripper_pos
+        extra_keys:
+        - left_joint_pos
+        - right_joint_pos
+        loss_weights: null
+      language:
+        delta_indices:
+        - 0
+        modality_keys:
+        - annotation.sub_task
+        normalization_mode: null
+        action_representation: null
+        exclude_state: false
+        action_type: null
+        action_format: null
+        normalize_rotation: true
+        wrist_keys: null
+        hand_keys: null
+        extra_keys: null
+        loss_weights: null
+    real_r1_pro_sharpa_relative_eef:
+      video:
+        delta_indices:
+        - -20
+        - 0
+        modality_keys:
+        - ego_view_res320x240_freq20
+        - left_wrist_view_res320x240_freq20
+        - right_wrist_view_res320x240_freq20
+        normalization_mode: null
+        action_representation: null
+        exclude_state: false
+        action_type: null
+        action_format: null
+        normalize_rotation: true
+        wrist_keys: null
+        hand_keys: null
+        extra_keys: null
+        loss_weights: null
+      state:
+        delta_indices:
+        - 0
+        modality_keys:
+        - left_wrist_eef
+        - right_wrist_eef
+        - left_hand_joints
+        - right_hand_joints
+        normalization_mode: null
+        action_representation: null
+        exclude_state: false
+        action_type: null
+        action_format: null
+        normalize_rotation: true
+        wrist_keys: null
+        hand_keys: null
+        extra_keys: null
+        loss_weights: null
+      action:
+        delta_indices:
+        - 0
+        - 1
+        - 2
+        - 3
+        - 4
+        - 5
+        - 6
+        - 7
+        - 8
+        - 9
+        - 10
+        - 11
+        - 12
+        - 13
+        - 14
+        - 15
+        - 16
+        - 17
+        - 18
+        - 19
+        - 20
+        - 21
+        - 22
+        - 23
+        - 24
+        - 25
+        - 26
+        - 27
+        - 28
+        - 29
+        - 30
+        - 31
+        - 32
+        - 33
+        - 34
+        - 35
+        - 36
+        - 37
+        - 38
+        - 39
+        modality_keys:
+        - left_wrist_eef
+        - right_wrist_eef
+        - left_hand_joints
+        - right_hand_joints
+        normalization_mode: null
+        action_representation:
+        - {}
+        - {}
+        - {}
+        - {}
+        exclude_state: false
+        action_type:
+        - {}
+        - {}
+        - {}
+        - {}
+        action_format:
+        - {}
+        - {}
+        - {}
+        - {}
+        normalize_rotation: true
+        wrist_keys:
+        - left_wrist_eef
+        - right_wrist_eef
+        hand_keys:
+        - left_hand_joints
+        - right_hand_joints
+        extra_keys: []
+        loss_weights: null
+      language:
+        delta_indices:
+        - 0
+        modality_keys:
+        - annotation.human.coarse_action
+        normalization_mode: null
+        action_representation: null
+        exclude_state: false
+        action_type: null
+        action_format: null
+        normalize_rotation: true
+        wrist_keys: null
+        hand_keys: null
+        extra_keys: null
+        loss_weights: null
+    real_r1_pro_sharpa_relative_eef_maxinsights:
+      video:
+        delta_indices:
+        - -30
+        - 0
+        modality_keys:
+        - ego_view_cropratio_res320x240_freq30
+        normalization_mode: null
+        action_representation: null
+        exclude_state: false
+        action_type: null
+        action_format: null
+        normalize_rotation: true
+        wrist_keys: null
+        hand_keys: null
+        extra_keys: null
+        loss_weights: null
+      state:
+        delta_indices:
+        - 0
+        modality_keys:
+        - left_wrist_eef
+        - right_wrist_eef
+        - left_hand_joints
+        - right_hand_joints
+        normalization_mode: null
+        action_representation: null
+        exclude_state: true
+        action_type: null
+        action_format: null
+        normalize_rotation: true
+        wrist_keys: null
+        hand_keys: null
+        extra_keys: null
+        loss_weights: null
+      action:
+        delta_indices:
+        - 0
+        - 1
+        - 2
+        - 3
+        - 4
+        - 5
+        - 6
+        - 7
+        - 8
+        - 9
+        - 10
+        - 11
+        - 12
+        - 13
+        - 14
+        - 15
+        - 16
+        - 17
+        - 18
+        - 19
+        - 20
+        - 21
+        - 22
+        - 23
+        - 24
+        - 25
+        - 26
+        - 27
+        - 28
+        - 29
+        - 30
+        - 31
+        - 32
+        - 33
+        - 34
+        - 35
+        - 36
+        - 37
+        - 38
+        - 39
+        modality_keys:
+        - left_wrist_eef
+        - right_wrist_eef
+        - left_hand_joints
+        - right_hand_joints
+        normalization_mode: null
+        action_representation:
+        - {}
+        - {}
+        - {}
+        - {}
+        exclude_state: false
+        action_type:
+        - {}
+        - {}
+        - {}
+        - {}
+        action_format:
+        - {}
+        - {}
+        - {}
+        - {}
+        normalize_rotation: true
+        wrist_keys:
+        - left_wrist_eef
+        - right_wrist_eef
+        hand_keys:
+        - left_hand_joints
+        - right_hand_joints
+        extra_keys: []
+        loss_weights: null
+      language:
+        delta_indices:
+        - 0
+        modality_keys:
+        - annotation.human.coarse_action
+        normalization_mode: null
+        action_representation: null
+        exclude_state: false
+        action_type: null
+        action_format: null
+        normalize_rotation: true
+        wrist_keys: null
+        hand_keys: null
+        extra_keys: null
+        loss_weights: null
+  download_cache: false
+  shard_size: 1024
+  episode_sampling_rate: 0.1
+  num_shards_per_epoch: 100000
+  override_pretraining_statistics: false
+  mode: single_turn
+  random_chop: 0.0
+  mock_dataset_mode: false
+  num_prompt_trajectories: 2
+  variable_num_demos: false
+  max_prompt_trajectories: 5
+  shuffle: true
+  seed: 24
+  subsample_ratio: 1.0
+  image_crop_size:
+  - 244
+  - 244
+  image_target_size:
+  - 224
+  - 224
+  video_backend: torchcodec
+training:
+  output_dir: nvidia/Cosmos-Reason2-2B
+  experiment_name: null
+  max_steps: 200000
+  global_batch_size: 1024
+  batch_size: 32
+  gradient_accumulation_steps: 1
+  use_muon: false
+  muon_lr: 0.005
+  use_legacy_wd_application: false
+  learning_rate: 5.0e-05
+  lr_scheduler_type: cosine
+  weight_decay: 1.0e-05
+  warmup_ratio: 0.05
+  warmup_steps: 0
+  max_grad_norm: 1.0
+  wsd_stable_ratio: 0.8
+  wsd_decay_type: cosine
+  optim: adamw_torch_fused
+  start_from_checkpoint: null
+  tf32: true
+  fp16: false
+  bf16: true
+  eval_bf16: true
+  logging_steps: 10
+  save_steps: 1000
+  save_total_limit: 5
+  save_vl_model: false
+  upload_checkpoints: true
+  upload_every: 1000
+  upload_last_n_checkpoints: 5
+  max_concurrent_uploads: 2
+  eval_strategy: 'no'
+  eval_steps: 500
+  eval_set_split_ratio: 0.1
+  eval_batch_size: 2
+  save_best_eval_metric_name: ''
+  save_best_eval_metric_greater_is_better: true
+  deepspeed_stage: 2
+  gradient_checkpointing: false
+  use_ddp: false
+  num_gpus: 256
+  dataloader_num_workers: 4
+  remove_unused_columns: false
+  use_wandb: true
+  wandb_project: human_pretraining_n15_galaxea_sharpa
+  enable_profiling: false
+  max_retries: 3
+  skip_spike: true
+  skip_spike_threshold: 5.0
+  skip_spike_ema_alpha: 0.99
+  skip_spike_max_consecutive: 10
+  assert_loss_less_than: null
+max_steps: 200000
+save_steps: 1000

experiment_cfg/config.yaml

@@ -0,0 +1,1341 @@
+!!python/object:groot.vla.omni.configs.base_config.Config
+data: !!python/object:groot.vla.omni.configs.data.data_config.DataConfig
+  datasets:
+  - !!python/object:groot.vla.omni.configs.data.data_config.SingleDatasetConfig
+    dataset_paths:
+    - /mnt/aws-lfs-02/shared/datasets/xdof.yam_v7_all_merged_global_task_exclude_bad_subtasks
+    dataset_type: physical_embodiment
+    embodiment_tag: xdof_relative_eef_relative_joint
+    mix_ratio: 0.1
+  - !!python/object:groot.vla.omni.configs.data.data_config.SingleDatasetConfig
+    dataset_paths:
+    - /mnt/aws-lfs-02/shared/datasets/xdof.yam_v7_subtask_only_merged_global_task
+    dataset_type: physical_embodiment
+    embodiment_tag: xdof_relative_eef_relative_joint_subtask
+    mix_ratio: 0.2
+  - !!python/object:groot.vla.omni.configs.data.data_config.SingleDatasetConfig
+    dataset_paths:
+    - /mnt/aws-lfs-02/shared/datasets/droid_101_success_idlefiltered_n17
+    - /mnt/aws-lfs-02/shared/datasets/droid_101_success_idlefiltered_n17_swapped
+    dataset_type: physical_embodiment
+    embodiment_tag: oxe_droid_relative_eef_relative_joint
+    mix_ratio: 0.1
+  - !!python/object:groot.vla.omni.configs.data.data_config.SingleDatasetConfig
+    dataset_paths:
+    - /mnt/aws-lfs-02/shared/datasets/galaxea_sharpa/real_g1.g1-in-the-wild-merged
+    dataset_type: physical_embodiment
+    embodiment_tag: real_g1_relative_eef_relative_joints
+    mix_ratio: 0.05
+  - !!python/object:groot.vla.omni.configs.data.data_config.SingleDatasetConfig
+    dataset_paths:
+    - /mnt/aws-lfs-02/shared/datasets/galaxea_sharpa/real_r1_pro_sharpa.inlab_play_real_robot_batch_1
+    - /mnt/aws-lfs-02/shared/datasets/galaxea_sharpa/real_r1_pro_sharpa.inlab_play_real_robot_batch_2
+    - /mnt/aws-lfs-02/shared/datasets/galaxea_sharpa/real_r1_pro_sharpa.miscellaneous_1k_trajectories
+    dataset_type: physical_embodiment
+    embodiment_tag: real_r1_pro_sharpa_relative_eef
+    mix_ratio: 0.05
+  - !!python/object:groot.vla.omni.configs.data.data_config.SingleDatasetConfig
+    dataset_paths:
+    - /mnt/aws-lfs-02/shared/datasets/galaxea_sharpa/mecka_lerobot/real_r1_pro_sharpa.mecka_batch1-2025-12-10-merged
+    - /mnt/aws-lfs-02/shared/datasets/galaxea_sharpa/mecka_lerobot/real_r1_pro_sharpa.mecka_batch3_2026-01-04-merged_backup
+    - /mnt/aws-lfs-02/shared/datasets/galaxea_sharpa/mecka_lerobot/real_r1_pro_sharpa.mecka_batch4_2026-01-05-merged_backup
+    - /mnt/aws-lfs-02/shared/datasets/galaxea_sharpa/mecka_lerobot/real_r1_pro_sharpa.mecka_batch5_2026-01-05-merged_backup
+    - /mnt/aws-lfs-02/shared/datasets/galaxea_sharpa/mecka_lerobot/real_r1_pro_sharpa.mecka_batch6_2026-01-05-merged_backup
+    - /mnt/aws-lfs-02/shared/datasets/galaxea_sharpa/mecka_lerobot/real_r1_pro_sharpa.mecka_batch10_2026-01-10-merged_backup
+    - /mnt/aws-lfs-02/shared/datasets/galaxea_sharpa/mecka_lerobot/real_r1_pro_sharpa.mecka_batch11_2026-01-10-merged_backup
+    - /mnt/aws-lfs-02/shared/datasets/galaxea_sharpa/mecka_lerobot/real_r1_pro_sharpa.mecka_batch12_2026-01-10-merged_backup
+    - /mnt/aws-lfs-02/shared/datasets/galaxea_sharpa/mecka_lerobot/real_r1_pro_sharpa.mecka_batch8_2026-01-10-merged_backup
+    - /mnt/aws-lfs-02/shared/datasets/galaxea_sharpa/mecka_lerobot/real_r1_pro_sharpa.mecka_batch9_2026-01-10-merged_backup
+    dataset_type: physical_embodiment
+    embodiment_tag: real_r1_pro_sharpa_relative_eef_mecka
+    mix_ratio: 0.25
+  - !!python/object:groot.vla.omni.configs.data.data_config.SingleDatasetConfig
+    dataset_paths:
+    - /mnt/aws-lfs-02/shared/datasets/galaxea_sharpa/maxinsights_lerobot_updated/1530hrs/real_r1_pro_sharpa.maxinsights_1530hrs_updated_train_set_merged
+    dataset_type: physical_embodiment
+    embodiment_tag: real_r1_pro_sharpa_relative_eef_maxinsights
+    mix_ratio: 0.2
+  - !!python/object:groot.vla.omni.configs.data.data_config.SingleDatasetConfig
+    dataset_paths:
+    - /mnt/aws-lfs-02/shared/datasets/galaxea_sharpa/real_r1_pro_sharpa.inlab_play_human_batch1
+    - /mnt/aws-lfs-02/shared/datasets/galaxea_sharpa/real_r1_pro_sharpa.inlab_play_human_batch2
+    - /mnt/aws-lfs-02/shared/datasets/galaxea_sharpa/real_r1_pro_sharpa.shirt_rolling_task24_2000_human_video_filter_n6_keep1619_demo_stats
+    - /mnt/aws-lfs-02/shared/datasets/galaxea_sharpa/real_r1_pro_sharpa.shirt_rolling_task15_2000_human_video_filter_n6_keep572_demo_stats
+    - /mnt/aws-lfs-02/shared/datasets/galaxea_sharpa/real_r1_pro_sharpa.sort_cards_human_filter_n6_keep523_demo_stats_overwrite_left_side_stats
+    - /mnt/aws-lfs-02/shared/datasets/galaxea_sharpa/real_r1_pro_sharpa.tong_task38_2000_human_video_overwrite_left_side_stats
+    - /mnt/aws-lfs-02/shared/datasets/galaxea_sharpa/real_r1_pro_sharpa.syringe_task30i_2000_human_video_filtered
+    - /mnt/aws-lfs-02/shared/datasets/galaxea_sharpa/real_r1_pro_sharpa.unscrew_bottle_task43_2000_human_video_fixed-duration
+    - /mnt/aws-lfs-02/shared/datasets/galaxea_sharpa/real_r1_pro_sharpa.unscrew_Jim_bottle_task47_600_human_video
+    - /mnt/aws-lfs-02/shared/datasets/galaxea_sharpa/real_r1_pro_sharpa.fold_shirt_task30b_500_human_video_halfdone
+    - /mnt/aws-lfs-02/shared/datasets/galaxea_sharpa/real_r1_pro_sharpa.fold_towel_task30c_500_human_video_halfdone
+    - /mnt/aws-lfs-02/shared/datasets/galaxea_sharpa/real_r1_pro_sharpa.sort_cards_task32e_1000_human_video
+    dataset_type: physical_embodiment
+    embodiment_tag: real_r1_pro_sharpa_relative_eef_human
+    mix_ratio: 0.05
+  download_cache: false
+  episode_sampling_rate: 0.1
+  image_crop_size:
+  - 244
+  - 244
+  image_target_size:
+  - 224
+  - 224
+  max_prompt_trajectories: 5
+  mock_dataset_mode: false
+  modality_configs:
+    oxe_droid_relative_eef_relative_joint:
+      action: !!python/object:groot.vla.omni.data.types.ModalityConfig
+        action_format:
+        - &id004 !!python/object/apply:groot.vla.omni.data.types.ActionFormat
+          - xyz+rot6d
+        - &id001 !!python/object/apply:groot.vla.omni.data.types.ActionFormat
+          - default
+        - *id001
+        action_representation:
+        - &id002 !!python/object/apply:groot.vla.omni.data.types.ActionRepresentation
+          - relative
+        - &id005 !!python/object/apply:groot.vla.omni.data.types.ActionRepresentation
+          - absolute
+        - *id002
+        action_type:
+        - &id006 !!python/object/apply:groot.vla.omni.data.types.ActionType
+          - eef
+        - &id003 !!python/object/apply:groot.vla.omni.data.types.ActionType
+          - non_eef
+        - *id003
+        delta_indices:
+        - 0
+        - 1
+        - 2
+        - 3
+        - 4
+        - 5
+        - 6
+        - 7
+        - 8
+        - 9
+        - 10
+        - 11
+        - 12
+        - 13
+        - 14
+        - 15
+        - 16
+        - 17
+        - 18
+        - 19
+        - 20
+        - 21
+        - 22
+        - 23
+        - 24
+        - 25
+        - 26
+        - 27
+        - 28
+        - 29
+        - 30
+        - 31
+        - 32
+        - 33
+        - 34
+        - 35
+        - 36
+        - 37
+        - 38
+        - 39
+        exclude_state: false
+        extra_keys:
+        - joint_position
+        hand_keys:
+        - gripper_position
+        loss_weights: null
+        modality_keys:
+        - eef_9d
+        - gripper_position
+        - joint_position
+        normalization_mode: null
+        normalize_rotation: true
+        wrist_keys:
+        - eef_9d
+      language: !!python/object:groot.vla.omni.data.types.ModalityConfig
+        action_format: null
+        action_representation: null
+        action_type: null
+        delta_indices:
+        - 0
+        exclude_state: false
+        extra_keys: null
+        hand_keys: null
+        loss_weights: null
+        modality_keys:
+        - annotation.language.language_instruction
+        - annotation.language.language_instruction_2
+        - annotation.language.language_instruction_3
+        normalization_mode: null
+        normalize_rotation: true
+        wrist_keys: null
+      state: !!python/object:groot.vla.omni.data.types.ModalityConfig
+        action_format: null
+        action_representation: null
+        action_type: null
+        delta_indices:
+        - 0
+        exclude_state: false
+        extra_keys: null
+        hand_keys: null
+        loss_weights: null
+        modality_keys:
+        - eef_9d
+        - gripper_position
+        - joint_position
+        normalization_mode: null
+        normalize_rotation: true
+        wrist_keys: null
+      video: !!python/object:groot.vla.omni.data.types.ModalityConfig
+        action_format: null
+        action_representation: null
+        action_type: null
+        delta_indices:
+        - -15
+        - 0
+        exclude_state: false
+        extra_keys: null
+        hand_keys: null
+        loss_weights: null
+        modality_keys:
+        - exterior_image_1_left
+        - wrist_image_left
+        normalization_mode: null
+        normalize_rotation: true
+        wrist_keys: null
+    real_g1_relative_eef_relative_joints:
+      action: !!python/object:groot.vla.omni.data.types.ModalityConfig
+        action_format:
+        - *id004
+        - *id004
+        - *id001
+        - *id001
+        - *id001
+        - *id001
+        - *id001
+        - *id001
+        - *id001
+        action_representation:
+        - *id002
+        - *id002
+        - *id005
+        - *id005
+        - *id002
+        - *id002
+        - *id005
+        - *id005
+        - *id005
+        action_type:
+        - *id006
+        - *id006
+        - *id003
+        - *id003
+        - *id003
+        - *id003
+        - *id003
+        - *id003
+        - *id003
+        delta_indices:
+        - 0
+        - 1
+        - 2
+        - 3
+        - 4
+        - 5
+        - 6
+        - 7
+        - 8
+        - 9
+        - 10
+        - 11
+        - 12
+        - 13
+        - 14
+        - 15
+        - 16
+        - 17
+        - 18
+        - 19
+        - 20
+        - 21
+        - 22
+        - 23
+        - 24
+        - 25
+        - 26
+        - 27
+        - 28
+        - 29
+        - 30
+        - 31
+        - 32
+        - 33
+        - 34
+        - 35
+        - 36
+        - 37
+        - 38
+        - 39
+        exclude_state: false
+        extra_keys:
+        - left_arm
+        - right_arm
+        - waist
+        - base_height_command
+        - navigate_command
+        hand_keys:
+        - left_hand
+        - right_hand
+        loss_weights: null
+        modality_keys:
+        - left_wrist_eef_9d
+        - right_wrist_eef_9d
+        - left_hand
+        - right_hand
+        - left_arm
+        - right_arm
+        - waist
+        - base_height_command
+        - navigate_command
+        normalization_mode: null
+        normalize_rotation: true
+        wrist_keys:
+        - left_wrist_eef_9d
+        - right_wrist_eef_9d
+      language: !!python/object:groot.vla.omni.data.types.ModalityConfig
+        action_format: null
+        action_representation: null
+        action_type: null
+        delta_indices:
+        - 0
+        exclude_state: false
+        extra_keys: null
+        hand_keys: null
+        loss_weights: null
+        modality_keys:
+        - annotation.human.task_description
+        normalization_mode: null
+        normalize_rotation: true
+        wrist_keys: null
+      state: !!python/object:groot.vla.omni.data.types.ModalityConfig
+        action_format: null
+        action_representation: null
+        action_type: null
+        delta_indices:
+        - 0
+        exclude_state: false
+        extra_keys: null
+        hand_keys: null
+        loss_weights: null
+        modality_keys:
+        - left_wrist_eef_9d
+        - right_wrist_eef_9d
+        - left_hand
+        - right_hand
+        - left_arm
+        - right_arm
+        - waist
+        normalization_mode: null
+        normalize_rotation: true
+        wrist_keys: null
+      video: !!python/object:groot.vla.omni.data.types.ModalityConfig
+        action_format: null
+        action_representation: null
+        action_type: null
+        delta_indices:
+        - -20
+        - 0
+        exclude_state: false
+        extra_keys: null
+        hand_keys: null
+        loss_weights: null
+        modality_keys:
+        - ego_view
+        normalization_mode: null
+        normalize_rotation: true
+        wrist_keys: null
+    real_r1_pro_sharpa_relative_eef:
+      action: !!python/object:groot.vla.omni.data.types.ModalityConfig
+        action_format:
+        - *id004
+        - *id004
+        - *id001
+        - *id001
+        action_representation:
+        - *id002
+        - *id002
+        - *id005
+        - *id005
+        action_type:
+        - *id006
+        - *id006
+        - *id003
+        - *id003
+        delta_indices:
+        - 0
+        - 1
+        - 2
+        - 3
+        - 4
+        - 5
+        - 6
+        - 7
+        - 8
+        - 9
+        - 10
+        - 11
+        - 12
+        - 13
+        - 14
+        - 15
+        - 16
+        - 17
+        - 18
+        - 19
+        - 20
+        - 21
+        - 22
+        - 23
+        - 24
+        - 25
+        - 26
+        - 27
+        - 28
+        - 29
+        - 30
+        - 31
+        - 32
+        - 33
+        - 34
+        - 35
+        - 36
+        - 37
+        - 38
+        - 39
+        exclude_state: false
+        extra_keys: []
+        hand_keys:
+        - left_hand_joints
+        - right_hand_joints
+        loss_weights: null
+        modality_keys:
+        - left_wrist_eef
+        - right_wrist_eef
+        - left_hand_joints
+        - right_hand_joints
+        normalization_mode: null
+        normalize_rotation: true
+        wrist_keys:
+        - left_wrist_eef
+        - right_wrist_eef
+      language: !!python/object:groot.vla.omni.data.types.ModalityConfig
+        action_format: null
+        action_representation: null
+        action_type: null
+        delta_indices:
+        - 0
+        exclude_state: false
+        extra_keys: null
+        hand_keys: null
+        loss_weights: null
+        modality_keys:
+        - annotation.human.coarse_action
+        normalization_mode: null
+        normalize_rotation: true
+        wrist_keys: null
+      state: !!python/object:groot.vla.omni.data.types.ModalityConfig
+        action_format: null
+        action_representation: null
+        action_type: null
+        delta_indices:
+        - 0
+        exclude_state: false
+        extra_keys: null
+        hand_keys: null
+        loss_weights: null
+        modality_keys:
+        - left_wrist_eef
+        - right_wrist_eef
+        - left_hand_joints
+        - right_hand_joints
+        normalization_mode: null
+        normalize_rotation: true
+        wrist_keys: null
+      video: !!python/object:groot.vla.omni.data.types.ModalityConfig
+        action_format: null
+        action_representation: null
+        action_type: null
+        delta_indices:
+        - -20
+        - 0
+        exclude_state: false
+        extra_keys: null
+        hand_keys: null
+        loss_weights: null
+        modality_keys:
+        - ego_view_res320x240_freq20
+        - left_wrist_view_res320x240_freq20
+        - right_wrist_view_res320x240_freq20
+        normalization_mode: null
+        normalize_rotation: true
+        wrist_keys: null
+    real_r1_pro_sharpa_relative_eef_human:
+      action: !!python/object:groot.vla.omni.data.types.ModalityConfig
+        action_format:
+        - *id004
+        - *id004
+        - *id001
+        - *id001
+        action_representation:
+        - *id002
+        - *id002
+        - *id005
+        - *id005
+        action_type:
+        - *id006
+        - *id006
+        - *id003
+        - *id003
+        delta_indices:
+        - 0
+        - 1
+        - 2
+        - 3
+        - 4
+        - 5
+        - 6
+        - 7
+        - 8
+        - 9
+        - 10
+        - 11
+        - 12
+        - 13
+        - 14
+        - 15
+        - 16
+        - 17
+        - 18
+        - 19
+        - 20
+        - 21
+        - 22
+        - 23
+        - 24
+        - 25
+        - 26
+        - 27
+        - 28
+        - 29
+        - 30
+        - 31
+        - 32
+        - 33
+        - 34
+        - 35
+        - 36
+        - 37
+        - 38
+        - 39
+        exclude_state: false
+        extra_keys: []
+        hand_keys:
+        - left_hand_joints
+        - right_hand_joints
+        loss_weights: null
+        modality_keys:
+        - left_wrist_eef
+        - right_wrist_eef
+        - left_hand_joints
+        - right_hand_joints
+        normalization_mode: null
+        normalize_rotation: true
+        wrist_keys:
+        - left_wrist_eef
+        - right_wrist_eef
+      language: !!python/object:groot.vla.omni.data.types.ModalityConfig
+        action_format: null
+        action_representation: null
+        action_type: null
+        delta_indices:
+        - 0
+        exclude_state: false
+        extra_keys: null
+        hand_keys: null
+        loss_weights: null
+        modality_keys:
+        - annotation.human.coarse_action
+        normalization_mode: null
+        normalize_rotation: true
+        wrist_keys: null
+      state: !!python/object:groot.vla.omni.data.types.ModalityConfig
+        action_format: null
+        action_representation: null
+        action_type: null
+        delta_indices:
+        - 0
+        exclude_state: true
+        extra_keys: null
+        hand_keys: null
+        loss_weights: null
+        modality_keys:
+        - left_wrist_eef
+        - right_wrist_eef
+        - left_hand_joints
+        - right_hand_joints
+        normalization_mode: null
+        normalize_rotation: true
+        wrist_keys: null
+      video: !!python/object:groot.vla.omni.data.types.ModalityConfig
+        action_format: null
+        action_representation: null
+        action_type: null
+        delta_indices:
+        - -20
+        - 0
+        exclude_state: false
+        extra_keys: null
+        hand_keys: null
+        loss_weights: null
+        modality_keys:
+        - ego_view_res320x240_freq20
+        - left_wrist_view_res320x240_freq20
+        - right_wrist_view_res320x240_freq20
+        normalization_mode: null
+        normalize_rotation: true
+        wrist_keys: null
+    real_r1_pro_sharpa_relative_eef_maxinsights:
+      action: !!python/object:groot.vla.omni.data.types.ModalityConfig
+        action_format:
+        - *id004
+        - *id004
+        - *id001
+        - *id001
+        action_representation:
+        - *id002
+        - *id002
+        - *id005
+        - *id005
+        action_type:
+        - *id006
+        - *id006
+        - *id003
+        - *id003
+        delta_indices:
+        - 0
+        - 1
+        - 2
+        - 3
+        - 4
+        - 5
+        - 6
+        - 7
+        - 8
+        - 9
+        - 10
+        - 11
+        - 12
+        - 13
+        - 14
+        - 15
+        - 16
+        - 17
+        - 18
+        - 19
+        - 20
+        - 21
+        - 22
+        - 23
+        - 24
+        - 25
+        - 26
+        - 27
+        - 28
+        - 29
+        - 30
+        - 31
+        - 32
+        - 33
+        - 34
+        - 35
+        - 36
+        - 37
+        - 38
+        - 39
+        exclude_state: false
+        extra_keys: []
+        hand_keys:
+        - left_hand_joints
+        - right_hand_joints
+        loss_weights: null
+        modality_keys:
+        - left_wrist_eef
+        - right_wrist_eef
+        - left_hand_joints
+        - right_hand_joints
+        normalization_mode: null
+        normalize_rotation: true
+        wrist_keys:
+        - left_wrist_eef
+        - right_wrist_eef
+      language: !!python/object:groot.vla.omni.data.types.ModalityConfig
+        action_format: null
+        action_representation: null
+        action_type: null
+        delta_indices:
+        - 0
+        exclude_state: false
+        extra_keys: null
+        hand_keys: null
+        loss_weights: null
+        modality_keys:
+        - annotation.human.coarse_action
+        normalization_mode: null
+        normalize_rotation: true
+        wrist_keys: null
+      state: !!python/object:groot.vla.omni.data.types.ModalityConfig
+        action_format: null
+        action_representation: null
+        action_type: null
+        delta_indices:
+        - 0
+        exclude_state: true
+        extra_keys: null
+        hand_keys: null
+        loss_weights: null
+        modality_keys:
+        - left_wrist_eef
+        - right_wrist_eef
+        - left_hand_joints
+        - right_hand_joints
+        normalization_mode: null
+        normalize_rotation: true
+        wrist_keys: null
+      video: !!python/object:groot.vla.omni.data.types.ModalityConfig
+        action_format: null
+        action_representation: null
+        action_type: null
+        delta_indices:
+        - -30
+        - 0
+        exclude_state: false
+        extra_keys: null
+        hand_keys: null
+        loss_weights: null
+        modality_keys:
+        - ego_view_cropratio_res320x240_freq30
+        normalization_mode: null
+        normalize_rotation: true
+        wrist_keys: null
+    real_r1_pro_sharpa_relative_eef_mecka:
+      action: !!python/object:groot.vla.omni.data.types.ModalityConfig
+        action_format:
+        - *id004
+        - *id004
+        - *id001
+        - *id001
+        action_representation:
+        - *id002
+        - *id002
+        - *id005
+        - *id005
+        action_type:
+        - *id006
+        - *id006
+        - *id003
+        - *id003
+        delta_indices:
+        - 0
+        - 1
+        - 2
+        - 3
+        - 4
+        - 5
+        - 6
+        - 7
+        - 8
+        - 9
+        - 10
+        - 11
+        - 12
+        - 13
+        - 14
+        - 15
+        - 16
+        - 17
+        - 18
+        - 19
+        - 20
+        - 21
+        - 22
+        - 23
+        - 24
+        - 25
+        - 26
+        - 27
+        - 28
+        - 29
+        - 30
+        - 31
+        - 32
+        - 33
+        - 34
+        - 35
+        - 36
+        - 37
+        - 38
+        - 39
+        exclude_state: false
+        extra_keys: []
+        hand_keys:
+        - left_hand_joints
+        - right_hand_joints
+        loss_weights: null
+        modality_keys:
+        - left_wrist_eef
+        - right_wrist_eef
+        - left_hand_joints
+        - right_hand_joints
+        normalization_mode: null
+        normalize_rotation: true
+        wrist_keys:
+        - left_wrist_eef
+        - right_wrist_eef
+      language: !!python/object:groot.vla.omni.data.types.ModalityConfig
+        action_format: null
+        action_representation: null
+        action_type: null
+        delta_indices:
+        - 0
+        exclude_state: false
+        extra_keys: null
+        hand_keys: null
+        loss_weights: null
+        modality_keys:
+        - annotation.human.coarse_action
+        normalization_mode: null
+        normalize_rotation: true
+        wrist_keys: null
+      state: !!python/object:groot.vla.omni.data.types.ModalityConfig
+        action_format: null
+        action_representation: null
+        action_type: null
+        delta_indices:
+        - 0
+        exclude_state: true
+        extra_keys: null
+        hand_keys: null
+        loss_weights: null
+        modality_keys:
+        - left_wrist_eef
+        - right_wrist_eef
+        - left_hand_joints
+        - right_hand_joints
+        normalization_mode: null
+        normalize_rotation: true
+        wrist_keys: null
+      video: !!python/object:groot.vla.omni.data.types.ModalityConfig
+        action_format: null
+        action_representation: null
+        action_type: null
+        delta_indices:
+        - -30
+        - 0
+        exclude_state: false
+        extra_keys: null
+        hand_keys: null
+        loss_weights: null
+        modality_keys:
+        - ego_view_cropratio_res320x240_freq30
+        normalization_mode: null
+        normalize_rotation: true
+        wrist_keys: null
+    xdof_relative_eef_relative_joint:
+      action: !!python/object:groot.vla.omni.data.types.ModalityConfig
+        action_format:
+        - *id004
+        - *id004
+        - *id001
+        - *id001
+        - *id001
+        - *id001
+        action_representation:
+        - *id002
+        - *id002
+        - *id005
+        - *id005
+        - *id002
+        - *id002
+        action_type:
+        - *id006
+        - *id006
+        - *id003
+        - *id003
+        - *id003
+        - *id003
+        delta_indices:
+        - 0
+        - 1
+        - 2
+        - 3
+        - 4
+        - 5
+        - 6
+        - 7
+        - 8
+        - 9
+        - 10
+        - 11
+        - 12
+        - 13
+        - 14
+        - 15
+        - 16
+        - 17
+        - 18
+        - 19
+        - 20
+        - 21
+        - 22
+        - 23
+        - 24
+        - 25
+        - 26
+        - 27
+        - 28
+        - 29
+        - 30
+        - 31
+        - 32
+        - 33
+        - 34
+        - 35
+        - 36
+        - 37
+        - 38
+        - 39
+        exclude_state: false
+        extra_keys:
+        - left_joint_pos
+        - right_joint_pos
+        hand_keys:
+        - left_gripper_pos
+        - right_gripper_pos
+        loss_weights: null
+        modality_keys:
+        - left_wrist_eef
+        - right_wrist_eef
+        - left_gripper_pos
+        - right_gripper_pos
+        - left_joint_pos
+        - right_joint_pos
+        normalization_mode: null
+        normalize_rotation: true
+        wrist_keys:
+        - left_wrist_eef
+        - right_wrist_eef
+      language: !!python/object:groot.vla.omni.data.types.ModalityConfig
+        action_format: null
+        action_representation: null
+        action_type: null
+        delta_indices:
+        - 0
+        exclude_state: false
+        extra_keys: null
+        hand_keys: null
+        loss_weights: null
+        modality_keys:
+        - annotation.task
+        normalization_mode: null
+        normalize_rotation: true
+        wrist_keys: null
+      state: !!python/object:groot.vla.omni.data.types.ModalityConfig
+        action_format: null
+        action_representation: null
+        action_type: null
+        delta_indices:
+        - 0
+        exclude_state: false
+        extra_keys: null
+        hand_keys: null
+        loss_weights: null
+        modality_keys:
+        - left_wrist_eef
+        - right_wrist_eef
+        - left_gripper_pos
+        - right_gripper_pos
+        - left_joint_pos
+        - right_joint_pos
+        normalization_mode: null
+        normalize_rotation: true
+        wrist_keys: null
+      video: !!python/object:groot.vla.omni.data.types.ModalityConfig
+        action_format: null
+        action_representation: null
+        action_type: null
+        delta_indices:
+        - -30
+        - 0
+        exclude_state: false
+        extra_keys: null
+        hand_keys: null
+        loss_weights: null
+        modality_keys:
+        - top_camera-images-rgb_320_240
+        - left_camera-images-rgb_320_240
+        - right_camera-images-rgb_320_240
+        normalization_mode: null
+        normalize_rotation: true
+        wrist_keys: null
+    xdof_relative_eef_relative_joint_subtask:
+      action: !!python/object:groot.vla.omni.data.types.ModalityConfig
+        action_format:
+        - *id004
+        - *id004
+        - *id001
+        - *id001
+        - *id001
+        - *id001
+        action_representation:
+        - *id002
+        - *id002
+        - *id005
+        - *id005
+        - *id002
+        - *id002
+        action_type:
+        - *id006
+        - *id006
+        - *id003
+        - *id003
+        - *id003
+        - *id003
+        delta_indices:
+        - 0
+        - 1
+        - 2
+        - 3
+        - 4
+        - 5
+        - 6
+        - 7
+        - 8
+        - 9
+        - 10
+        - 11
+        - 12
+        - 13
+        - 14
+        - 15
+        - 16
+        - 17
+        - 18
+        - 19
+        - 20
+        - 21
+        - 22
+        - 23
+        - 24
+        - 25
+        - 26
+        - 27
+        - 28
+        - 29
+        - 30
+        - 31
+        - 32
+        - 33
+        - 34
+        - 35
+        - 36
+        - 37
+        - 38
+        - 39
+        exclude_state: false
+        extra_keys:
+        - left_joint_pos
+        - right_joint_pos
+        hand_keys:
+        - left_gripper_pos
+        - right_gripper_pos
+        loss_weights: null
+        modality_keys:
+        - left_wrist_eef
+        - right_wrist_eef
+        - left_gripper_pos
+        - right_gripper_pos
+        - left_joint_pos
+        - right_joint_pos
+        normalization_mode: null
+        normalize_rotation: true
+        wrist_keys:
+        - left_wrist_eef
+        - right_wrist_eef
+      language: !!python/object:groot.vla.omni.data.types.ModalityConfig
+        action_format: null
+        action_representation: null
+        action_type: null
+        delta_indices:
+        - 0
+        exclude_state: false
+        extra_keys: null
+        hand_keys: null
+        loss_weights: null
+        modality_keys:
+        - annotation.sub_task
+        normalization_mode: null
+        normalize_rotation: true
+        wrist_keys: null
+      state: !!python/object:groot.vla.omni.data.types.ModalityConfig
+        action_format: null
+        action_representation: null
+        action_type: null
+        delta_indices:
+        - 0
+        exclude_state: false
+        extra_keys: null
+        hand_keys: null
+        loss_weights: null
+        modality_keys:
+        - left_wrist_eef
+        - right_wrist_eef
+        - left_gripper_pos
+        - right_gripper_pos
+        - left_joint_pos
+        - right_joint_pos
+        normalization_mode: null
+        normalize_rotation: true
+        wrist_keys: null
+      video: !!python/object:groot.vla.omni.data.types.ModalityConfig
+        action_format: null
+        action_representation: null
+        action_type: null
+        delta_indices:
+        - -30
+        - 0
+        exclude_state: false
+        extra_keys: null
+        hand_keys: null
+        loss_weights: null
+        modality_keys:
+        - top_camera-images-rgb_320_240
+        - left_camera-images-rgb_320_240
+        - right_camera-images-rgb_320_240
+        normalization_mode: null
+        normalize_rotation: true
+        wrist_keys: null
+  mode: single_turn
+  num_prompt_trajectories: 2
+  num_shards_per_epoch: 100000
+  override_pretraining_statistics: false
+  random_chop: 0.0
+  seed: 24
+  shard_size: 1024
+  shuffle: true
+  subsample_ratio: 1.0
+  variable_num_demos: false
+  video_backend: torchcodec
+load_config_path: groot/vla/omni/configs/experiments/r1_pro/sharpa/n17_pretrain/n17_pretrain_human_robot_cross_embodiment_fix_yam_absolute_hand_2step.yaml
+model: !!python/object:groot.vla.omni.configs.model.groot_n1d5_qwen.GrootN1d5QwenConfig
+  _attn_implementation_internal: null
+  _commit_hash: null
+  _name_or_path: ''
+  _output_attentions: false
+  action_horizon: 40
+  action_space_prompt: false
+  add_cross_attention: false
+  add_pos_embed: true
+  apply_sincos_state_encoding: false
+  architectures: null
+  attn_dropout: 0.2
+  backbone_embedding_dim: 2048
+  bad_words_ids: null
+  begin_suppress_tokens: null
+  bos_token_id: null
+  chunk_size_feed_forward: 0
+  color_jitter_params:
+    brightness: 0.3
+    contrast: 0.4
+    hue: 0.08
+    saturation: 0.5
+  crop_fraction: 0.95
+  cross_attention_hidden_size: null
+  decoder_start_token_id: null
+  diffusion_model_cfg:
+    attention_head_dim: 48
+    cross_attention_dim: 2048
+    dropout: 0.2
+    final_dropout: true
+    interleave_self_attention: true
+    norm_type: ada_norm
+    num_attention_heads: 32
+    num_layers: 32
+    output_dim: 1024
+    positional_embeddings: null
+  dit_latent_dim: 1536
+  diversity_penalty: 0.0
+  do_human_interpolation: false
+  do_sample: false
+  dtype: null
+  early_stopping: false
+  encoder_no_repeat_ngram_size: 0
+  eos_token_id: null
+  exclude_state: false
+  exponential_decay_length_penalty: null
+  finetuning_task: null
+  forced_bos_token_id: null
+  forced_eos_token_id: null
+  formalize_language: true
+  hidden_size: 1024
+  human_embodiment_tags: null
+  id2label:
+    0: LABEL_0
+    1: LABEL_1
+  image_crop_size: !!python/tuple
+  - 230
+  - 230
+  image_target_size: !!python/tuple
+  - 256
+  - 256
+  interpolation_steps: 20
+  is_decoder: false
+  is_encoder_decoder: false
+  label2id:
+    LABEL_0: 0
+    LABEL_1: 1
+  language_dropout_prob: 0.0
+  length_penalty: 1.0
+  letter_box_transform: false
+  load_bf16: true
+  max_action_dim: 132
+  max_length: 20
+  max_num_embodiments: 32
+  max_seq_len: 1024
+  max_state_dim: 132
+  min_length: 0
+  model_dtype: bfloat16
+  model_type: GrootN1d5Qwen
+  no_repeat_ngram_size: 0
+  noise_beta_alpha: 1.5
+  noise_beta_beta: 1.0
+  noise_s: 0.999
+  num_beam_groups: 1
+  num_beams: 1
+  num_inference_timesteps: 4
+  num_return_sequences: 1
+  num_timestep_buckets: 1000
+  output_hidden_states: false
+  output_scores: false
+  pad_token_id: null
+  prefix: null
+  problem_type: null
+  pruned_heads: {}
+  random_history_crop: true
+  random_rotation_angle: 0
+  remove_invalid_values: false
+  repetition_penalty: 1.0
+  reproject_vision: false
+  return_dict: true
+  return_dict_in_generate: false
+  rtc_ramp_rate: 6.0
+  select_layer: 16
+  sep_token_id: null
+  shortest_image_edge: 256
+  state_dropout_prob: 0.2
+  state_gaussian_noise_std: 0.0
+  suppress_tokens: null
+  task_specific_params: null
+  temperature: 1.0
+  tf_legacy_loss: false
+  tie_encoder_decoder: false
+  tie_word_embeddings: true
+  tokenizer_class: null
+  top_k: 50
+  top_p: 1.0
+  torchscript: false
+  transformers_version: null
+  tune_diffusion_model: true
+  tune_linear: true
+  tune_llm: false
+  tune_projector: true
+  tune_top_llm_layers: 0
+  tune_visual: false
+  tune_vlln: true
+  typical_p: 1.0
+  use_albumentations: true
+  use_alternate_vl_dit: true
+  use_bfloat16: false
+  use_flash_attention: true
+  use_future_tokens: false
+  use_mean_std: false
+  use_percentiles: true
+  use_vl_self_attention: true
+  use_vlln: true
+  vl_self_attention_cfg:
+    attention_head_dim: 64
+    dropout: 0.2
+    final_dropout: true
+    num_attention_heads: 32
+    num_layers: 4
+    positional_embeddings: null
+  vlm_backend: qwen3
+  vlm_model_path: nvidia/Cosmos-Reason2-2B
+training: !!python/object:groot.vla.omni.configs.training.training_config.TrainingConfig
+  assert_loss_less_than: null
+  batch_size: 32
+  bf16: true
+  dataloader_num_workers: 4
+  deepspeed_stage: 2
+  enable_profiling: false
+  eval_batch_size: 2
+  eval_bf16: true
+  eval_set_split_ratio: 0.1
+  eval_steps: 500
+  eval_strategy: 'no'
+  experiment_name: null
+  fp16: false
+  global_batch_size: 1024
+  gradient_accumulation_steps: 1
+  gradient_checkpointing: false
+  learning_rate: 5.0e-05
+  logging_steps: 10
+  lr_scheduler_type: cosine
+  max_concurrent_uploads: 2
+  max_grad_norm: 1.0
+  max_retries: 3
+  max_steps: 200000
+  muon_lr: 0.005
+  num_gpus: 256
+  optim: adamw_torch_fused
+  output_dir: nvidia/Cosmos-Reason2-2B
+  remove_unused_columns: false
+  save_best_eval_metric_greater_is_better: true
+  save_best_eval_metric_name: ''
+  save_steps: 1000
+  save_total_limit: 5
+  save_vl_model: false
+  skip_spike: true
+  skip_spike_ema_alpha: 0.99
+  skip_spike_max_consecutive: 10
+  skip_spike_threshold: 5.0
+  start_from_checkpoint: null
+  tf32: true
+  upload_checkpoints: true
+  upload_every: 1000
+  upload_last_n_checkpoints: 5
+  use_ddp: false
+  use_legacy_wd_application: false
+  use_muon: false
+  use_wandb: true
+  wandb_project: human_pretraining_n15_galaxea_sharpa
+  warmup_ratio: 0.05
+  warmup_steps: 0
+  weight_decay: 1.0e-05
+  wsd_decay_type: cosine
+  wsd_stable_ratio: 0.8

experiment_cfg/final_model_config.json

@@ -0,0 +1,56 @@
+{
+  "model_type": "Gr00tN1d7",
+  "model_dtype": "bfloat16",
+  "model_name": "nvidia/Cosmos-Reason2-2B",
+  "backbone_model_type": "qwen",
+  "model_revision": null,
+  "tune_top_llm_layers": 4,
+  "backbone_embedding_dim": 2048,
+  "tune_llm": true,
+  "tune_visual": true,
+  "select_layer": 16,
+  "reproject_vision": false,
+  "use_flash_attention": true,
+  "load_bf16": true,
+  "collator_overwrite_image_inputs": false,
+  "eagle_collator": false,
+  "backbone_trainable_params_fp32": true,
+  "gemma_collator": false,
+  "apply_sincos_state_encoding": true,
+  "use_percentiles": false,
+  "use_relative_action": true,
+  "max_state_dim": 128,
+  "max_action_dim": 128,
+  "action_horizon": 50,
+  "hidden_size": 1024,
+  "input_embedding_dim": 1536,
+  "state_history_length": 1,
+  "add_pos_embed": true,
+  "attn_dropout": 0.2,
+  "use_vlln": true,
+  "max_seq_len": 1024,
+  "use_alternate_vl_dit": true,
+  "attend_text_every_n_blocks": 2,
+  "diffusion_model_cfg": {
+    "positional_embeddings": null,
+    "num_layers": 32,
+    "num_attention_heads": 32,
+    "attention_head_dim": 48,
+    "norm_type": "ada_norm",
+    "dropout": 0.2,
+    "final_dropout": true,
+    "output_dim": 1024,
+    "interleave_self_attention": true
+  },
+  "num_inference_timesteps": 4,
+  "noise_beta_alpha": 1.5,
+  "noise_beta_beta": 1.0,
+  "noise_s": 0.999,
+  "num_timestep_buckets": 1000,
+  "tune_projector": true,
+  "tune_diffusion_model": true,
+  "tune_vlln": true,
+  "state_dropout_prob": 0.0,
+  "state_additive_noise_scale": 0.0,
+  "max_num_embodiments": 32
+}

experiment_cfg/final_processor_config.json

@@ -0,0 +1,27 @@
+{
+  "modality_configs": "{'xdof': {'video': ModalityConfig(delta_indices=[0], modality_keys=['left_camera-images-rgb_320_240', 'top_camera-images-rgb_320_240', 'right_camera-images-rgb_320_240'], sin_cos_embedding_keys=None, mean_std_embedding_keys=None, action_configs=None), 'state': ModalityConfig(delta_indices=[0], modality_keys=['gripper_pos_obs_left', 'gripper_pos_obs_right', 'joint_pos_obs_left', 'joint_pos_obs_right'], sin_cos_embedding_keys=None, mean_std_embedding_keys=None, action_configs=None), 'action': ModalityConfig(delta_indices=[0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49], modality_keys=['gripper_pos_action_left', 'gripper_pos_action_right', 'joint_pos_action_left', 'joint_pos_action_right'], sin_cos_embedding_keys=None, mean_std_embedding_keys=None, action_configs=[ActionConfig(rep=<ActionRepresentation.ABSOLUTE: 'absolute'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key='gripper_pos_obs_left'), ActionConfig(rep=<ActionRepresentation.ABSOLUTE: 'absolute'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key='gripper_pos_obs_right'), ActionConfig(rep=<ActionRepresentation.RELATIVE: 'relative'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key='joint_pos_obs_left'), ActionConfig(rep=<ActionRepresentation.RELATIVE: 'relative'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key='joint_pos_obs_right')]), 'language': ModalityConfig(delta_indices=[0], modality_keys=['annotation.task'], sin_cos_embedding_keys=None, mean_std_embedding_keys=None, action_configs=None)}, 'unitree_g1_full_body_with_waist_height_nav_cmd': {'video': ModalityConfig(delta_indices=[0], modality_keys=['ego_view'], sin_cos_embedding_keys=None, mean_std_embedding_keys=None, action_configs=None), 'state': ModalityConfig(delta_indices=[0], modality_keys=['left_leg', 'right_leg', 'waist', 'left_arm', 'right_arm', 'left_hand', 'right_hand'], sin_cos_embedding_keys=None, mean_std_embedding_keys=None, action_configs=None), 'action': ModalityConfig(delta_indices=[0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49], modality_keys=['left_arm', 'right_arm', 'left_hand', 'right_hand', 'waist', 'base_height_command', 'navigate_command'], sin_cos_embedding_keys=None, mean_std_embedding_keys=None, action_configs=[ActionConfig(rep=<ActionRepresentation.RELATIVE: 'relative'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None), ActionConfig(rep=<ActionRepresentation.RELATIVE: 'relative'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None), ActionConfig(rep=<ActionRepresentation.ABSOLUTE: 'absolute'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None), ActionConfig(rep=<ActionRepresentation.ABSOLUTE: 'absolute'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None), ActionConfig(rep=<ActionRepresentation.ABSOLUTE: 'absolute'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None), ActionConfig(rep=<ActionRepresentation.ABSOLUTE: 'absolute'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None), ActionConfig(rep=<ActionRepresentation.ABSOLUTE: 'absolute'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None)]), 'language': ModalityConfig(delta_indices=[0], modality_keys=['annotation.human.task_description'], sin_cos_embedding_keys=None, mean_std_embedding_keys=None, action_configs=None)}, 'oxe_droid_joint_position_relative': {'video': ModalityConfig(delta_indices=[0], modality_keys=['exterior_image_1_left', 'wrist_image_left'], sin_cos_embedding_keys=None, mean_std_embedding_keys=None, action_configs=None), 'state': ModalityConfig(delta_indices=[0], modality_keys=['joint_position', 'gripper_position'], sin_cos_embedding_keys=None, mean_std_embedding_keys=None, action_configs=None), 'action': ModalityConfig(delta_indices=[0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31], modality_keys=['joint_position', 'gripper_position'], sin_cos_embedding_keys=None, mean_std_embedding_keys=None, action_configs=[ActionConfig(rep=<ActionRepresentation.RELATIVE: 'relative'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None), ActionConfig(rep=<ActionRepresentation.ABSOLUTE: 'absolute'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None)]), 'language': ModalityConfig(delta_indices=[0], modality_keys=['annotation.language.language_instruction', 'annotation.language.language_instruction_2', 'annotation.language.language_instruction_3'], sin_cos_embedding_keys=None, mean_std_embedding_keys=None, action_configs=None)}, 'simpler_env_google': {'video': ModalityConfig(delta_indices=[0], modality_keys=['image'], sin_cos_embedding_keys=None, mean_std_embedding_keys=None, action_configs=None), 'state': ModalityConfig(delta_indices=[0], modality_keys=['x', 'y', 'z', 'rx', 'ry', 'rz', 'rw', 'gripper'], sin_cos_embedding_keys=None, mean_std_embedding_keys=None, action_configs=None), 'action': ModalityConfig(delta_indices=[0, 1, 2, 3, 4, 5, 6, 7], modality_keys=['x', 'y', 'z', 'roll', 'pitch', 'yaw', 'gripper'], sin_cos_embedding_keys=None, mean_std_embedding_keys=['x', 'y', 'z', 'roll', 'pitch', 'yaw'], action_configs=[ActionConfig(rep=<ActionRepresentation.ABSOLUTE: 'absolute'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None), ActionConfig(rep=<ActionRepresentation.ABSOLUTE: 'absolute'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None), ActionConfig(rep=<ActionRepresentation.ABSOLUTE: 'absolute'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None), ActionConfig(rep=<ActionRepresentation.ABSOLUTE: 'absolute'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None), ActionConfig(rep=<ActionRepresentation.ABSOLUTE: 'absolute'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None), ActionConfig(rep=<ActionRepresentation.ABSOLUTE: 'absolute'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None), ActionConfig(rep=<ActionRepresentation.ABSOLUTE: 'absolute'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None)]), 'language': ModalityConfig(delta_indices=[0], modality_keys=['annotation.human.action.task_description'], sin_cos_embedding_keys=None, mean_std_embedding_keys=None, action_configs=None)}, 'robocasa_panda_omron': {'video': ModalityConfig(delta_indices=[0], modality_keys=['res256_image_side_0', 'res256_image_side_1', 'res256_image_wrist_0'], sin_cos_embedding_keys=None, mean_std_embedding_keys=None, action_configs=None), 'state': ModalityConfig(delta_indices=[0], modality_keys=['end_effector_position_relative', 'end_effector_rotation_relative', 'gripper_qpos', 'base_position', 'base_rotation'], sin_cos_embedding_keys=None, mean_std_embedding_keys=None, action_configs=None), 'action': ModalityConfig(delta_indices=[0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15], modality_keys=['end_effector_position', 'end_effector_rotation', 'gripper_close', 'base_motion', 'control_mode'], sin_cos_embedding_keys=None, mean_std_embedding_keys=None, action_configs=[ActionConfig(rep=<ActionRepresentation.ABSOLUTE: 'absolute'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None), ActionConfig(rep=<ActionRepresentation.ABSOLUTE: 'absolute'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None), ActionConfig(rep=<ActionRepresentation.ABSOLUTE: 'absolute'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None), ActionConfig(rep=<ActionRepresentation.ABSOLUTE: 'absolute'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None), ActionConfig(rep=<ActionRepresentation.ABSOLUTE: 'absolute'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None)]), 'language': ModalityConfig(delta_indices=[0], modality_keys=['annotation.human.action.task_description'], sin_cos_embedding_keys=None, mean_std_embedding_keys=None, action_configs=None)}, 'gr1_unified': {'video': ModalityConfig(delta_indices=[0], modality_keys=['ego_view_bg_crop_pad_res256_freq20'], sin_cos_embedding_keys=None, mean_std_embedding_keys=None, action_configs=None), 'state': ModalityConfig(delta_indices=[0], modality_keys=['left_arm', 'right_arm', 'left_hand', 'right_hand', 'waist'], sin_cos_embedding_keys=['left_arm', 'right_arm', 'left_hand', 'right_hand', 'waist'], mean_std_embedding_keys=None, action_configs=None), 'action': ModalityConfig(delta_indices=[0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15], modality_keys=['left_arm', 'right_arm', 'left_hand', 'right_hand', 'waist'], sin_cos_embedding_keys=None, mean_std_embedding_keys=None, action_configs=[ActionConfig(rep=<ActionRepresentation.RELATIVE: 'relative'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None), ActionConfig(rep=<ActionRepresentation.RELATIVE: 'relative'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None), ActionConfig(rep=<ActionRepresentation.RELATIVE: 'relative'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None), ActionConfig(rep=<ActionRepresentation.RELATIVE: 'relative'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None), ActionConfig(rep=<ActionRepresentation.ABSOLUTE: 'absolute'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None)]), 'language': ModalityConfig(delta_indices=[0], modality_keys=['task'], sin_cos_embedding_keys=None, mean_std_embedding_keys=None, action_configs=None), 'rl_info': ModalityConfig(delta_indices=[0], modality_keys=[], sin_cos_embedding_keys=None, mean_std_embedding_keys=None, action_configs=None)}, 'sim_behavior_r1_pro': {'video': ModalityConfig(delta_indices=[0], modality_keys=['observation.images.rgb.head_256_256', 'observation.images.rgb.left_wrist_256_256', 'observation.images.rgb.right_wrist_256_256'], sin_cos_embedding_keys=None, mean_std_embedding_keys=None, action_configs=None), 'state': ModalityConfig(delta_indices=[0], modality_keys=['robot_pos', 'robot_ori_cos', 'robot_ori_sin', 'robot_2d_ori', 'robot_2d_ori_cos', 'robot_2d_ori_sin', 'robot_lin_vel', 'robot_ang_vel', 'arm_left_qpos', 'arm_left_qpos_sin', 'arm_left_qpos_cos', 'eef_left_pos', 'eef_left_quat', 'gripper_left_qpos', 'arm_right_qpos', 'arm_right_qpos_sin', 'arm_right_qpos_cos', 'eef_right_pos', 'eef_right_quat', 'gripper_right_qpos', 'trunk_qpos'], sin_cos_embedding_keys=None, mean_std_embedding_keys=None, action_configs=None), 'action': ModalityConfig(delta_indices=[0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31], modality_keys=['base', 'torso', 'left_arm', 'left_gripper', 'right_arm', 'right_gripper'], sin_cos_embedding_keys=None, mean_std_embedding_keys=None, action_configs=[ActionConfig(rep=<ActionRepresentation.ABSOLUTE: 'absolute'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None), ActionConfig(rep=<ActionRepresentation.RELATIVE: 'relative'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key='trunk_qpos'), ActionConfig(rep=<ActionRepresentation.RELATIVE: 'relative'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key='arm_left_qpos'), ActionConfig(rep=<ActionRepresentation.ABSOLUTE: 'absolute'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None), ActionConfig(rep=<ActionRepresentation.RELATIVE: 'relative'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key='arm_right_qpos'), ActionConfig(rep=<ActionRepresentation.ABSOLUTE: 'absolute'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None)]), 'language': ModalityConfig(delta_indices=[0], modality_keys=['annotation.human.coarse_action'], sin_cos_embedding_keys=None, mean_std_embedding_keys=None, action_configs=None)}, 'simpler_env_widowx': {'video': ModalityConfig(delta_indices=[0], modality_keys=['image_0'], sin_cos_embedding_keys=None, mean_std_embedding_keys=None, action_configs=None), 'state': ModalityConfig(delta_indices=[0], modality_keys=['x', 'y', 'z', 'roll', 'pitch', 'yaw', 'pad', 'gripper'], sin_cos_embedding_keys=None, mean_std_embedding_keys=None, action_configs=None), 'action': ModalityConfig(delta_indices=[0, 1, 2, 3, 4, 5, 6, 7], modality_keys=['x', 'y', 'z', 'roll', 'pitch', 'yaw', 'gripper'], sin_cos_embedding_keys=None, mean_std_embedding_keys=['x', 'y', 'z', 'roll', 'pitch', 'yaw'], action_configs=[ActionConfig(rep=<ActionRepresentation.ABSOLUTE: 'absolute'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None), ActionConfig(rep=<ActionRepresentation.ABSOLUTE: 'absolute'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None), ActionConfig(rep=<ActionRepresentation.ABSOLUTE: 'absolute'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None), ActionConfig(rep=<ActionRepresentation.ABSOLUTE: 'absolute'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None), ActionConfig(rep=<ActionRepresentation.ABSOLUTE: 'absolute'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None), ActionConfig(rep=<ActionRepresentation.ABSOLUTE: 'absolute'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None), ActionConfig(rep=<ActionRepresentation.ABSOLUTE: 'absolute'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None)]), 'language': ModalityConfig(delta_indices=[0], modality_keys=['annotation.human.action.task_description'], sin_cos_embedding_keys=None, mean_std_embedding_keys=None, action_configs=None)}, 'agibot': {'video': ModalityConfig(delta_indices=[0], modality_keys=['top_head_pad_res256_freq10', 'hand_left_pad_res256_freq10', 'hand_right_pad_res256_freq10'], sin_cos_embedding_keys=None, mean_std_embedding_keys=None, action_configs=None), 'state': ModalityConfig(delta_indices=[0], modality_keys=['left_arm_joint_position', 'right_arm_joint_position', 'left_effector_position', 'right_effector_position', 'head_position', 'waist_pitch', 'waist_lift'], sin_cos_embedding_keys=['left_arm_joint_position', 'right_arm_joint_position', 'head_position', 'waist_pitch'], mean_std_embedding_keys=None, action_configs=None), 'action': ModalityConfig(delta_indices=[0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49], modality_keys=['left_arm_joint_position', 'right_arm_joint_position', 'left_effector_position', 'right_effector_position', 'head_position', 'waist_pitch', 'waist_lift', 'robot_velocity'], sin_cos_embedding_keys=None, mean_std_embedding_keys=None, action_configs=[ActionConfig(rep=<ActionRepresentation.RELATIVE: 'relative'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None), ActionConfig(rep=<ActionRepresentation.RELATIVE: 'relative'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None), ActionConfig(rep=<ActionRepresentation.ABSOLUTE: 'absolute'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None), ActionConfig(rep=<ActionRepresentation.ABSOLUTE: 'absolute'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None), ActionConfig(rep=<ActionRepresentation.ABSOLUTE: 'absolute'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None), ActionConfig(rep=<ActionRepresentation.ABSOLUTE: 'absolute'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None), ActionConfig(rep=<ActionRepresentation.ABSOLUTE: 'absolute'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None), ActionConfig(rep=<ActionRepresentation.ABSOLUTE: 'absolute'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None)]), 'language': ModalityConfig(delta_indices=[0], modality_keys=['annotation.language.action_text'], sin_cos_embedding_keys=None, mean_std_embedding_keys=None, action_configs=None)}}",
+  "state_action_processor": "StateActionProcessor(modality_configs={'xdof': {'video': ModalityConfig(delta_indices=[0], modality_keys=['left_camera-images-rgb_320_240', 'top_camera-images-rgb_320_240', 'right_camera-images-rgb_320_240'], sin_cos_embedding_keys=None, mean_std_embedding_keys=None, action_configs=None), 'state': ModalityConfig(delta_indices=[0], modality_keys=['gripper_pos_obs_left', 'gripper_pos_obs_right', 'joint_pos_obs_left', 'joint_pos_obs_right'], sin_cos_embedding_keys=None, mean_std_embedding_keys=None, action_configs=None), 'action': ModalityConfig(delta_indices=[0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49], modality_keys=['gripper_pos_action_left', 'gripper_pos_action_right', 'joint_pos_action_left', 'joint_pos_action_right'], sin_cos_embedding_keys=None, mean_std_embedding_keys=None, action_configs=[ActionConfig(rep=<ActionRepresentation.ABSOLUTE: 'absolute'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key='gripper_pos_obs_left'), ActionConfig(rep=<ActionRepresentation.ABSOLUTE: 'absolute'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key='gripper_pos_obs_right'), ActionConfig(rep=<ActionRepresentation.RELATIVE: 'relative'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key='joint_pos_obs_left'), ActionConfig(rep=<ActionRepresentation.RELATIVE: 'relative'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key='joint_pos_obs_right')]), 'language': ModalityConfig(delta_indices=[0], modality_keys=['annotation.task'], sin_cos_embedding_keys=None, mean_std_embedding_keys=None, action_configs=None)}, 'unitree_g1_full_body_with_waist_height_nav_cmd': {'video': ModalityConfig(delta_indices=[0], modality_keys=['ego_view'], sin_cos_embedding_keys=None, mean_std_embedding_keys=None, action_configs=None), 'state': ModalityConfig(delta_indices=[0], modality_keys=['left_leg', 'right_leg', 'waist', 'left_arm', 'right_arm', 'left_hand', 'right_hand'], sin_cos_embedding_keys=None, mean_std_embedding_keys=None, action_configs=None), 'action': ModalityConfig(delta_indices=[0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49], modality_keys=['left_arm', 'right_arm', 'left_hand', 'right_hand', 'waist', 'base_height_command', 'navigate_command'], sin_cos_embedding_keys=None, mean_std_embedding_keys=None, action_configs=[ActionConfig(rep=<ActionRepresentation.RELATIVE: 'relative'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None), ActionConfig(rep=<ActionRepresentation.RELATIVE: 'relative'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None), ActionConfig(rep=<ActionRepresentation.ABSOLUTE: 'absolute'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None), ActionConfig(rep=<ActionRepresentation.ABSOLUTE: 'absolute'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None), ActionConfig(rep=<ActionRepresentation.ABSOLUTE: 'absolute'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None), ActionConfig(rep=<ActionRepresentation.ABSOLUTE: 'absolute'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None), ActionConfig(rep=<ActionRepresentation.ABSOLUTE: 'absolute'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None)]), 'language': ModalityConfig(delta_indices=[0], modality_keys=['annotation.human.task_description'], sin_cos_embedding_keys=None, mean_std_embedding_keys=None, action_configs=None)}, 'oxe_droid_joint_position_relative': {'video': ModalityConfig(delta_indices=[0], modality_keys=['exterior_image_1_left', 'wrist_image_left'], sin_cos_embedding_keys=None, mean_std_embedding_keys=None, action_configs=None), 'state': ModalityConfig(delta_indices=[0], modality_keys=['joint_position', 'gripper_position'], sin_cos_embedding_keys=None, mean_std_embedding_keys=None, action_configs=None), 'action': ModalityConfig(delta_indices=[0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31], modality_keys=['joint_position', 'gripper_position'], sin_cos_embedding_keys=None, mean_std_embedding_keys=None, action_configs=[ActionConfig(rep=<ActionRepresentation.RELATIVE: 'relative'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None), ActionConfig(rep=<ActionRepresentation.ABSOLUTE: 'absolute'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None)]), 'language': ModalityConfig(delta_indices=[0], modality_keys=['annotation.language.language_instruction', 'annotation.language.language_instruction_2', 'annotation.language.language_instruction_3'], sin_cos_embedding_keys=None, mean_std_embedding_keys=None, action_configs=None)}, 'simpler_env_google': {'video': ModalityConfig(delta_indices=[0], modality_keys=['image'], sin_cos_embedding_keys=None, mean_std_embedding_keys=None, action_configs=None), 'state': ModalityConfig(delta_indices=[0], modality_keys=['x', 'y', 'z', 'rx', 'ry', 'rz', 'rw', 'gripper'], sin_cos_embedding_keys=None, mean_std_embedding_keys=None, action_configs=None), 'action': ModalityConfig(delta_indices=[0, 1, 2, 3, 4, 5, 6, 7], modality_keys=['x', 'y', 'z', 'roll', 'pitch', 'yaw', 'gripper'], sin_cos_embedding_keys=None, mean_std_embedding_keys=['x', 'y', 'z', 'roll', 'pitch', 'yaw'], action_configs=[ActionConfig(rep=<ActionRepresentation.ABSOLUTE: 'absolute'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None), ActionConfig(rep=<ActionRepresentation.ABSOLUTE: 'absolute'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None), ActionConfig(rep=<ActionRepresentation.ABSOLUTE: 'absolute'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None), ActionConfig(rep=<ActionRepresentation.ABSOLUTE: 'absolute'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None), ActionConfig(rep=<ActionRepresentation.ABSOLUTE: 'absolute'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None), ActionConfig(rep=<ActionRepresentation.ABSOLUTE: 'absolute'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None), ActionConfig(rep=<ActionRepresentation.ABSOLUTE: 'absolute'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None)]), 'language': ModalityConfig(delta_indices=[0], modality_keys=['annotation.human.action.task_description'], sin_cos_embedding_keys=None, mean_std_embedding_keys=None, action_configs=None)}, 'robocasa_panda_omron': {'video': ModalityConfig(delta_indices=[0], modality_keys=['res256_image_side_0', 'res256_image_side_1', 'res256_image_wrist_0'], sin_cos_embedding_keys=None, mean_std_embedding_keys=None, action_configs=None), 'state': ModalityConfig(delta_indices=[0], modality_keys=['end_effector_position_relative', 'end_effector_rotation_relative', 'gripper_qpos', 'base_position', 'base_rotation'], sin_cos_embedding_keys=None, mean_std_embedding_keys=None, action_configs=None), 'action': ModalityConfig(delta_indices=[0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15], modality_keys=['end_effector_position', 'end_effector_rotation', 'gripper_close', 'base_motion', 'control_mode'], sin_cos_embedding_keys=None, mean_std_embedding_keys=None, action_configs=[ActionConfig(rep=<ActionRepresentation.ABSOLUTE: 'absolute'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None), ActionConfig(rep=<ActionRepresentation.ABSOLUTE: 'absolute'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None), ActionConfig(rep=<ActionRepresentation.ABSOLUTE: 'absolute'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None), ActionConfig(rep=<ActionRepresentation.ABSOLUTE: 'absolute'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None), ActionConfig(rep=<ActionRepresentation.ABSOLUTE: 'absolute'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None)]), 'language': ModalityConfig(delta_indices=[0], modality_keys=['annotation.human.action.task_description'], sin_cos_embedding_keys=None, mean_std_embedding_keys=None, action_configs=None)}, 'gr1_unified': {'video': ModalityConfig(delta_indices=[0], modality_keys=['ego_view_bg_crop_pad_res256_freq20'], sin_cos_embedding_keys=None, mean_std_embedding_keys=None, action_configs=None), 'state': ModalityConfig(delta_indices=[0], modality_keys=['left_arm', 'right_arm', 'left_hand', 'right_hand', 'waist'], sin_cos_embedding_keys=['left_arm', 'right_arm', 'left_hand', 'right_hand', 'waist'], mean_std_embedding_keys=None, action_configs=None), 'action': ModalityConfig(delta_indices=[0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15], modality_keys=['left_arm', 'right_arm', 'left_hand', 'right_hand', 'waist'], sin_cos_embedding_keys=None, mean_std_embedding_keys=None, action_configs=[ActionConfig(rep=<ActionRepresentation.RELATIVE: 'relative'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None), ActionConfig(rep=<ActionRepresentation.RELATIVE: 'relative'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None), ActionConfig(rep=<ActionRepresentation.RELATIVE: 'relative'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None), ActionConfig(rep=<ActionRepresentation.RELATIVE: 'relative'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None), ActionConfig(rep=<ActionRepresentation.ABSOLUTE: 'absolute'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None)]), 'language': ModalityConfig(delta_indices=[0], modality_keys=['task'], sin_cos_embedding_keys=None, mean_std_embedding_keys=None, action_configs=None), 'rl_info': ModalityConfig(delta_indices=[0], modality_keys=[], sin_cos_embedding_keys=None, mean_std_embedding_keys=None, action_configs=None)}, 'sim_behavior_r1_pro': {'video': ModalityConfig(delta_indices=[0], modality_keys=['observation.images.rgb.head_256_256', 'observation.images.rgb.left_wrist_256_256', 'observation.images.rgb.right_wrist_256_256'], sin_cos_embedding_keys=None, mean_std_embedding_keys=None, action_configs=None), 'state': ModalityConfig(delta_indices=[0], modality_keys=['robot_pos', 'robot_ori_cos', 'robot_ori_sin', 'robot_2d_ori', 'robot_2d_ori_cos', 'robot_2d_ori_sin', 'robot_lin_vel', 'robot_ang_vel', 'arm_left_qpos', 'arm_left_qpos_sin', 'arm_left_qpos_cos', 'eef_left_pos', 'eef_left_quat', 'gripper_left_qpos', 'arm_right_qpos', 'arm_right_qpos_sin', 'arm_right_qpos_cos', 'eef_right_pos', 'eef_right_quat', 'gripper_right_qpos', 'trunk_qpos'], sin_cos_embedding_keys=None, mean_std_embedding_keys=None, action_configs=None), 'action': ModalityConfig(delta_indices=[0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31], modality_keys=['base', 'torso', 'left_arm', 'left_gripper', 'right_arm', 'right_gripper'], sin_cos_embedding_keys=None, mean_std_embedding_keys=None, action_configs=[ActionConfig(rep=<ActionRepresentation.ABSOLUTE: 'absolute'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None), ActionConfig(rep=<ActionRepresentation.RELATIVE: 'relative'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key='trunk_qpos'), ActionConfig(rep=<ActionRepresentation.RELATIVE: 'relative'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key='arm_left_qpos'), ActionConfig(rep=<ActionRepresentation.ABSOLUTE: 'absolute'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None), ActionConfig(rep=<ActionRepresentation.RELATIVE: 'relative'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key='arm_right_qpos'), ActionConfig(rep=<ActionRepresentation.ABSOLUTE: 'absolute'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None)]), 'language': ModalityConfig(delta_indices=[0], modality_keys=['annotation.human.coarse_action'], sin_cos_embedding_keys=None, mean_std_embedding_keys=None, action_configs=None)}, 'simpler_env_widowx': {'video': ModalityConfig(delta_indices=[0], modality_keys=['image_0'], sin_cos_embedding_keys=None, mean_std_embedding_keys=None, action_configs=None), 'state': ModalityConfig(delta_indices=[0], modality_keys=['x', 'y', 'z', 'roll', 'pitch', 'yaw', 'pad', 'gripper'], sin_cos_embedding_keys=None, mean_std_embedding_keys=None, action_configs=None), 'action': ModalityConfig(delta_indices=[0, 1, 2, 3, 4, 5, 6, 7], modality_keys=['x', 'y', 'z', 'roll', 'pitch', 'yaw', 'gripper'], sin_cos_embedding_keys=None, mean_std_embedding_keys=['x', 'y', 'z', 'roll', 'pitch', 'yaw'], action_configs=[ActionConfig(rep=<ActionRepresentation.ABSOLUTE: 'absolute'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None), ActionConfig(rep=<ActionRepresentation.ABSOLUTE: 'absolute'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None), ActionConfig(rep=<ActionRepresentation.ABSOLUTE: 'absolute'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None), ActionConfig(rep=<ActionRepresentation.ABSOLUTE: 'absolute'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None), ActionConfig(rep=<ActionRepresentation.ABSOLUTE: 'absolute'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None), ActionConfig(rep=<ActionRepresentation.ABSOLUTE: 'absolute'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None), ActionConfig(rep=<ActionRepresentation.ABSOLUTE: 'absolute'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None)]), 'language': ModalityConfig(delta_indices=[0], modality_keys=['annotation.human.action.task_description'], sin_cos_embedding_keys=None, mean_std_embedding_keys=None, action_configs=None)}, 'agibot': {'video': ModalityConfig(delta_indices=[0], modality_keys=['top_head_pad_res256_freq10', 'hand_left_pad_res256_freq10', 'hand_right_pad_res256_freq10'], sin_cos_embedding_keys=None, mean_std_embedding_keys=None, action_configs=None), 'state': ModalityConfig(delta_indices=[0], modality_keys=['left_arm_joint_position', 'right_arm_joint_position', 'left_effector_position', 'right_effector_position', 'head_position', 'waist_pitch', 'waist_lift'], sin_cos_embedding_keys=['left_arm_joint_position', 'right_arm_joint_position', 'head_position', 'waist_pitch'], mean_std_embedding_keys=None, action_configs=None), 'action': ModalityConfig(delta_indices=[0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49], modality_keys=['left_arm_joint_position', 'right_arm_joint_position', 'left_effector_position', 'right_effector_position', 'head_position', 'waist_pitch', 'waist_lift', 'robot_velocity'], sin_cos_embedding_keys=None, mean_std_embedding_keys=None, action_configs=[ActionConfig(rep=<ActionRepresentation.RELATIVE: 'relative'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None), ActionConfig(rep=<ActionRepresentation.RELATIVE: 'relative'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None), ActionConfig(rep=<ActionRepresentation.ABSOLUTE: 'absolute'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None), ActionConfig(rep=<ActionRepresentation.ABSOLUTE: 'absolute'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None), ActionConfig(rep=<ActionRepresentation.ABSOLUTE: 'absolute'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None), ActionConfig(rep=<ActionRepresentation.ABSOLUTE: 'absolute'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None), ActionConfig(rep=<ActionRepresentation.ABSOLUTE: 'absolute'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None), ActionConfig(rep=<ActionRepresentation.ABSOLUTE: 'absolute'>, type=<ActionType.NON_EEF: 'non_eef'>, format=<ActionFormat.DEFAULT: 'default'>, state_key=None)]), 'language': ModalityConfig(delta_indices=[0], modality_keys=['annotation.language.action_text'], sin_cos_embedding_keys=None, mean_std_embedding_keys=None, action_configs=None)}}, statistics={}, use_percentiles=False, clip_outliers=True, apply_sincos_state_encoding=True, use_relative_action=True)",
+  "use_percentiles": "False",
+  "clip_outliers": "True",
+  "apply_sincos_state_encoding": "True",
+  "use_relative_action": "True",
+  "formalize_language": "True",
+  "model_name": "nvidia/Cosmos-Reason2-2B",
+  "model_type": "qwen",
+  "processor": "Qwen3VLProcessor:\n- image_processor: Qwen2VLImageProcessorFast {\n  \"crop_size\": null,\n  \"data_format\": \"channels_first\",\n  \"default_to_square\": true,\n  \"device\": null,\n  \"disable_grouping\": null,\n  \"do_center_crop\": null,\n  \"do_convert_rgb\": true,\n  \"do_normalize\": true,\n  \"do_pad\": null,\n  \"do_rescale\": true,\n  \"do_resize\": true,\n  \"image_mean\": [\n    0.5,\n    0.5,\n    0.5\n  ],\n  \"image_processor_type\": \"Qwen2VLImageProcessorFast\",\n  \"image_std\": [\n    0.5,\n    0.5,\n    0.5\n  ],\n  \"input_data_format\": null,\n  \"max_pixels\": null,\n  \"merge_size\": 2,\n  \"min_pixels\": null,\n  \"pad_size\": null,\n  \"patch_size\": 16,\n  \"processor_class\": \"Qwen3VLProcessor\",\n  \"resample\": 3,\n  \"rescale_factor\": 0.00392156862745098,\n  \"return_tensors\": null,\n  \"size\": {\n    \"longest_edge\": 16777216,\n    \"shortest_edge\": 65536\n  },\n  \"temporal_patch_size\": 2\n}\n\n- tokenizer: Qwen2TokenizerFast(name_or_path='nvidia/Cosmos-Reason2-2B', vocab_size=151643, model_max_length=262144, is_fast=True, padding_side='left', truncation_side='right', special_tokens={'eos_token': '<|im_end|>', 'pad_token': '<|endoftext|>', 'additional_special_tokens': ['<|im_start|>', '<|im_end|>', '<|object_ref_start|>', '<|object_ref_end|>', '<|box_start|>', '<|box_end|>', '<|quad_start|>', '<|quad_end|>', '<|vision_start|>', '<|vision_end|>', '<|vision_pad|>', '<|image_pad|>', '<|video_pad|>']}, clean_up_tokenization_spaces=False, added_tokens_decoder={\n\t151643: AddedToken(\"<|endoftext|>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True),\n\t151644: AddedToken(\"<|im_start|>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True),\n\t151645: AddedToken(\"<|im_end|>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True),\n\t151646: AddedToken(\"<|object_ref_start|>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True),\n\t151647: AddedToken(\"<|object_ref_end|>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True),\n\t151648: AddedToken(\"<|box_start|>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True),\n\t151649: AddedToken(\"<|box_end|>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True),\n\t151650: AddedToken(\"<|quad_start|>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True),\n\t151651: AddedToken(\"<|quad_end|>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True),\n\t151652: AddedToken(\"<|vision_start|>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True),\n\t151653: AddedToken(\"<|vision_end|>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True),\n\t151654: AddedToken(\"<|vision_pad|>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True),\n\t151655: AddedToken(\"<|image_pad|>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True),\n\t151656: AddedToken(\"<|video_pad|>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=True),\n\t151657: AddedToken(\"<tool_call>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=False),\n\t151658: AddedToken(\"</tool_call>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=False),\n\t151659: AddedToken(\"<|fim_prefix|>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=False),\n\t151660: AddedToken(\"<|fim_middle|>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=False),\n\t151661: AddedToken(\"<|fim_suffix|>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=False),\n\t151662: AddedToken(\"<|fim_pad|>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=False),\n\t151663: AddedToken(\"<|repo_name|>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=False),\n\t151664: AddedToken(\"<|file_sep|>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=False),\n\t151665: AddedToken(\"<tool_response>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=False),\n\t151666: AddedToken(\"</tool_response>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=False),\n\t151667: AddedToken(\"<think>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=False),\n\t151668: AddedToken(\"</think>\", rstrip=False, lstrip=False, single_word=False, normalized=False, special=False),\n}\n)\n- video_processor: Qwen3VLVideoProcessor {\n  \"crop_size\": null,\n  \"data_format\": \"channels_first\",\n  \"default_to_square\": true,\n  \"device\": null,\n  \"do_center_crop\": null,\n  \"do_convert_rgb\": true,\n  \"do_normalize\": true,\n  \"do_rescale\": true,\n  \"do_resize\": true,\n  \"do_sample_frames\": true,\n  \"fps\": 2,\n  \"image_mean\": [\n    0.5,\n    0.5,\n    0.5\n  ],\n  \"image_std\": [\n    0.5,\n    0.5,\n    0.5\n  ],\n  \"input_data_format\": null,\n  \"max_frames\": 768,\n  \"merge_size\": 2,\n  \"min_frames\": 4,\n  \"num_frames\": null,\n  \"pad_size\": null,\n  \"patch_size\": 16,\n  \"processor_class\": \"Qwen3VLProcessor\",\n  \"resample\": 3,\n  \"rescale_factor\": 0.00392156862745098,\n  \"return_metadata\": false,\n  \"size\": {\n    \"longest_edge\": 25165824,\n    \"shortest_edge\": 4096\n  },\n  \"temporal_patch_size\": 2,\n  \"video_metadata\": null,\n  \"video_processor_type\": \"Qwen3VLVideoProcessor\"\n}\n\n\n{\n  \"processor_class\": \"Qwen3VLProcessor\"\n}\n",
+  "max_state_dim": "128",
+  "max_action_dim": "128",
+  "max_action_horizon": "50",
+  "image_crop_size": "None",
+  "image_target_size": "None",
+  "random_rotation_angle": "None",
+  "color_jitter_params": "{'brightness': 0.1, 'contrast': 0.1, 'saturation': 0.1, 'hue': 0.1}",
+  "embodiment_id_mapping": "{'robocasa_panda_omron': 13, 'oxe_droid': 17, 'oxe_droid_joint_position_relative': 17, 'oxe_fractal': 18, 'oxe_language_table': 19, 'oxe_bridge': 20, 'unknown': 22, 'gr1_unified': 20, 'agibot': 26, 'oxe_mutex': 28, 'oxe_roboset': 29, 'oxe_plex': 30, 'dream': 31, 'language_table_sim': 7, 'gr1_isaac': 0, 'xdof': 23, 'xdof_oss_data': 27, 'xdof_h16': 23, 'sim_behavior_r1_pro': 24, 'unitree_g1_full_body_with_waist_height_nav_cmd': 25, 'unitree_g1_full_body_with_waist_height_nav_cmd_sim': 8, 'unitree_g1_whole_body_teleop_latent': 9, 'unitree_g1_whole_body_teleop_smpl': 16, 'simpler_env_google': 0, 'simpler_env_widowx': 1, 'libero_sim': 2, 'droid_sim': 3, 'real_r1_pro_sharpa': 8, 'r1_pro': 27, 'r1_pro_single-view': 27, 'new_embodiment': 10, 'so100_2rgb': 6, 'so100_3rgb': 6, 'robomind_agilex_3rgb': 4, 'robomind_franka_1rgb': 5, 'robomind_franka_3rgb': 5, 'robomind_tienkung_gello_1rgb': 11, 'robomind_ur_1rgb': 12, 'robomind_tienkung_xsens_1rgb': 13, 'molmoact_franka_3rgb': 14, 'galaxea_r1_4rgb': 15}",
+  "shortest_image_edge": "256",
+  "crop_fraction": "0.95",
+  "use_albumentations": "True",
+  "train_image_transform": "ReplayCompose([\n  SmallestMaxSize(p=1.0, max_size=[256], interpolation=3),\n  FractionalRandomCrop(p=1.0, crop_fraction=0.95),\n  SmallestMaxSize(p=1.0, max_size=[256], interpolation=3),\n  ColorJitter(p=1.0, brightness=(0.9, 1.1), contrast=(0.9, 1.1), saturation=(0.9, 1.1), hue=(-0.1, 0.1)),\n], p=1.0, bbox_params=None, keypoint_params=None, additional_targets={}, is_check_shapes=True, save_key='replay')",
+  "eval_image_transform": "Compose([\n  SmallestMaxSize(p=1.0, max_size=[256], interpolation=3),\n  FractionalCenterCrop(p=1.0, crop_fraction=0.95),\n  SmallestMaxSize(p=1.0, max_size=[256], interpolation=3),\n], p=1.0, bbox_params=None, keypoint_params=None, additional_targets={}, is_check_shapes=True)",
+  "_collator": "Gr00tN1d7DataCollator(model_name=nvidia/Cosmos-Reason2-2B, model_type=qwen)",
+  "training": "True"
+}

experiment_cfg/initial_actions.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c34d9e522cf105de4e5c8b2c3d6024cb1e40cdbfcf7c64c7ff9b0fc738c6f932
+size 1564431

latest

@@ -0,0 +1 @@
+global_step196000

model-00001-of-00002.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8a1a1d8a33c99103c7c80c136073c5bb8bfe9ca8f7a970c93c033ea89742906d
+size 4990519232

model-00002-of-00002.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c3f61940deb2007ba1ad7743013b57f0f8462356151db9655175d7aca2d40661
+size 1919980184

processor_config.json

@@ -0,0 +1,955 @@
+{
+  "processor_class": "Gr00tN1d7Processor",
+  "processor_kwargs": {
+    "modality_configs": {
+      "real_g1_relative_eef_relative_joints": {
+        "video": {
+          "delta_indices": [
+            -20,
+            0
+          ],
+          "modality_keys": [
+            "ego_view"
+          ]
+        },
+        "state": {
+          "delta_indices": [
+            0
+          ],
+          "modality_keys": [
+            "left_wrist_eef_9d",
+            "right_wrist_eef_9d",
+            "left_hand",
+            "right_hand",
+            "left_arm",
+            "right_arm",
+            "waist"
+          ]
+        },
+        "action": {
+          "delta_indices": [
+            0,
+            1,
+            2,
+            3,
+            4,
+            5,
+            6,
+            7,
+            8,
+            9,
+            10,
+            11,
+            12,
+            13,
+            14,
+            15,
+            16,
+            17,
+            18,
+            19,
+            20,
+            21,
+            22,
+            23,
+            24,
+            25,
+            26,
+            27,
+            28,
+            29,
+            30,
+            31,
+            32,
+            33,
+            34,
+            35,
+            36,
+            37,
+            38,
+            39
+          ],
+          "modality_keys": [
+            "left_wrist_eef_9d",
+            "right_wrist_eef_9d",
+            "left_hand",
+            "right_hand",
+            "left_arm",
+            "right_arm",
+            "waist",
+            "base_height_command",
+            "navigate_command"
+          ],
+          "action_configs": [
+            {
+              "rep": "RELATIVE",
+              "type": "EEF",
+              "format": "XYZ_ROT6D",
+              "state_key": "left_wrist_eef_9d"
+            },
+            {
+              "rep": "RELATIVE",
+              "type": "EEF",
+              "format": "XYZ_ROT6D",
+              "state_key": "right_wrist_eef_9d"
+            },
+            {
+              "rep": "ABSOLUTE",
+              "type": "NON_EEF",
+              "format": "DEFAULT",
+              "state_key": "left_hand"
+            },
+            {
+              "rep": "ABSOLUTE",
+              "type": "NON_EEF",
+              "format": "DEFAULT",
+              "state_key": "right_hand"
+            },
+            {
+              "rep": "RELATIVE",
+              "type": "NON_EEF",
+              "format": "DEFAULT",
+              "state_key": "left_arm"
+            },
+            {
+              "rep": "RELATIVE",
+              "type": "NON_EEF",
+              "format": "DEFAULT",
+              "state_key": "right_arm"
+            },
+            {
+              "rep": "ABSOLUTE",
+              "type": "NON_EEF",
+              "format": "DEFAULT",
+              "state_key": "waist"
+            },
+            {
+              "rep": "ABSOLUTE",
+              "type": "NON_EEF",
+              "format": "DEFAULT",
+              "state_key": "base_height_command"
+            },
+            {
+              "rep": "ABSOLUTE",
+              "type": "NON_EEF",
+              "format": "DEFAULT",
+              "state_key": "navigate_command"
+            }
+          ]
+        },
+        "language": {
+          "delta_indices": [
+            0
+          ],
+          "modality_keys": [
+            "annotation.human.task_description"
+          ]
+        }
+      },
+      "real_r1_pro_sharpa_relative_eef_mecka": {
+        "video": {
+          "delta_indices": [
+            -30,
+            0
+          ],
+          "modality_keys": [
+            "ego_view_cropratio_res320x240_freq30"
+          ]
+        },
+        "state": {
+          "delta_indices": [
+            0
+          ],
+          "modality_keys": [
+            "left_wrist_eef",
+            "right_wrist_eef",
+            "left_hand_joints",
+            "right_hand_joints"
+          ]
+        },
+        "action": {
+          "delta_indices": [
+            0,
+            1,
+            2,
+            3,
+            4,
+            5,
+            6,
+            7,
+            8,
+            9,
+            10,
+            11,
+            12,
+            13,
+            14,
+            15,
+            16,
+            17,
+            18,
+            19,
+            20,
+            21,
+            22,
+            23,
+            24,
+            25,
+            26,
+            27,
+            28,
+            29,
+            30,
+            31,
+            32,
+            33,
+            34,
+            35,
+            36,
+            37,
+            38,
+            39
+          ],
+          "modality_keys": [
+            "left_wrist_eef",
+            "right_wrist_eef",
+            "left_hand_joints",
+            "right_hand_joints"
+          ],
+          "action_configs": [
+            {
+              "rep": "RELATIVE",
+              "type": "EEF",
+              "format": "XYZ_ROT6D",
+              "state_key": "left_wrist_eef"
+            },
+            {
+              "rep": "RELATIVE",
+              "type": "EEF",
+              "format": "XYZ_ROT6D",
+              "state_key": "right_wrist_eef"
+            },
+            {
+              "rep": "ABSOLUTE",
+              "type": "NON_EEF",
+              "format": "DEFAULT",
+              "state_key": "left_hand_joints"
+            },
+            {
+              "rep": "ABSOLUTE",
+              "type": "NON_EEF",
+              "format": "DEFAULT",
+              "state_key": "right_hand_joints"
+            }
+          ]
+        },
+        "language": {
+          "delta_indices": [
+            0
+          ],
+          "modality_keys": [
+            "annotation.human.coarse_action"
+          ]
+        }
+      },
+      "real_r1_pro_sharpa_relative_eef_human": {
+        "video": {
+          "delta_indices": [
+            -20,
+            0
+          ],
+          "modality_keys": [
+            "ego_view_res320x240_freq20",
+            "left_wrist_view_res320x240_freq20",
+            "right_wrist_view_res320x240_freq20"
+          ]
+        },
+        "state": {
+          "delta_indices": [
+            0
+          ],
+          "modality_keys": [
+            "left_wrist_eef",
+            "right_wrist_eef",
+            "left_hand_joints",
+            "right_hand_joints"
+          ]
+        },
+        "action": {
+          "delta_indices": [
+            0,
+            1,
+            2,
+            3,
+            4,
+            5,
+            6,
+            7,
+            8,
+            9,
+            10,
+            11,
+            12,
+            13,
+            14,
+            15,
+            16,
+            17,
+            18,
+            19,
+            20,
+            21,
+            22,
+            23,
+            24,
+            25,
+            26,
+            27,
+            28,
+            29,
+            30,
+            31,
+            32,
+            33,
+            34,
+            35,
+            36,
+            37,
+            38,
+            39
+          ],
+          "modality_keys": [
+            "left_wrist_eef",
+            "right_wrist_eef",
+            "left_hand_joints",
+            "right_hand_joints"
+          ],
+          "action_configs": [
+            {
+              "rep": "RELATIVE",
+              "type": "EEF",
+              "format": "XYZ_ROT6D",
+              "state_key": "left_wrist_eef"
+            },
+            {
+              "rep": "RELATIVE",
+              "type": "EEF",
+              "format": "XYZ_ROT6D",
+              "state_key": "right_wrist_eef"
+            },
+            {
+              "rep": "ABSOLUTE",
+              "type": "NON_EEF",
+              "format": "DEFAULT",
+              "state_key": "left_hand_joints"
+            },
+            {
+              "rep": "ABSOLUTE",
+              "type": "NON_EEF",
+              "format": "DEFAULT",
+              "state_key": "right_hand_joints"
+            }
+          ]
+        },
+        "language": {
+          "delta_indices": [
+            0
+          ],
+          "modality_keys": [
+            "annotation.human.coarse_action"
+          ]
+        }
+      },
+      "real_r1_pro_sharpa_relative_eef": {
+        "video": {
+          "delta_indices": [
+            -20,
+            0
+          ],
+          "modality_keys": [
+            "ego_view_res320x240_freq20",
+            "left_wrist_view_res320x240_freq20",
+            "right_wrist_view_res320x240_freq20"
+          ]
+        },
+        "state": {
+          "delta_indices": [
+            0
+          ],
+          "modality_keys": [
+            "left_wrist_eef",
+            "right_wrist_eef",
+            "left_hand_joints",
+            "right_hand_joints"
+          ]
+        },
+        "action": {
+          "delta_indices": [
+            0,
+            1,
+            2,
+            3,
+            4,
+            5,
+            6,
+            7,
+            8,
+            9,
+            10,
+            11,
+            12,
+            13,
+            14,
+            15,
+            16,
+            17,
+            18,
+            19,
+            20,
+            21,
+            22,
+            23,
+            24,
+            25,
+            26,
+            27,
+            28,
+            29,
+            30,
+            31,
+            32,
+            33,
+            34,
+            35,
+            36,
+            37,
+            38,
+            39
+          ],
+          "modality_keys": [
+            "left_wrist_eef",
+            "right_wrist_eef",
+            "left_hand_joints",
+            "right_hand_joints"
+          ],
+          "action_configs": [
+            {
+              "rep": "RELATIVE",
+              "type": "EEF",
+              "format": "XYZ_ROT6D",
+              "state_key": "left_wrist_eef"
+            },
+            {
+              "rep": "RELATIVE",
+              "type": "EEF",
+              "format": "XYZ_ROT6D",
+              "state_key": "right_wrist_eef"
+            },
+            {
+              "rep": "ABSOLUTE",
+              "type": "NON_EEF",
+              "format": "DEFAULT",
+              "state_key": "left_hand_joints"
+            },
+            {
+              "rep": "ABSOLUTE",
+              "type": "NON_EEF",
+              "format": "DEFAULT",
+              "state_key": "right_hand_joints"
+            }
+          ]
+        },
+        "language": {
+          "delta_indices": [
+            0
+          ],
+          "modality_keys": [
+            "annotation.human.coarse_action"
+          ]
+        }
+      },
+      "xdof_relative_eef_relative_joint": {
+        "video": {
+          "delta_indices": [
+            -30,
+            0
+          ],
+          "modality_keys": [
+            "top_camera-images-rgb_320_240",
+            "left_camera-images-rgb_320_240",
+            "right_camera-images-rgb_320_240"
+          ]
+        },
+        "state": {
+          "delta_indices": [
+            0
+          ],
+          "modality_keys": [
+            "left_wrist_eef",
+            "right_wrist_eef",
+            "left_gripper_pos",
+            "right_gripper_pos",
+            "left_joint_pos",
+            "right_joint_pos"
+          ]
+        },
+        "action": {
+          "delta_indices": [
+            0,
+            1,
+            2,
+            3,
+            4,
+            5,
+            6,
+            7,
+            8,
+            9,
+            10,
+            11,
+            12,
+            13,
+            14,
+            15,
+            16,
+            17,
+            18,
+            19,
+            20,
+            21,
+            22,
+            23,
+            24,
+            25,
+            26,
+            27,
+            28,
+            29,
+            30,
+            31,
+            32,
+            33,
+            34,
+            35,
+            36,
+            37,
+            38,
+            39
+          ],
+          "modality_keys": [
+            "left_wrist_eef",
+            "right_wrist_eef",
+            "left_gripper_pos",
+            "right_gripper_pos",
+            "left_joint_pos",
+            "right_joint_pos"
+          ],
+          "action_configs": [
+            {
+              "rep": "RELATIVE",
+              "type": "EEF",
+              "format": "XYZ_ROT6D",
+              "state_key": "left_wrist_eef"
+            },
+            {
+              "rep": "RELATIVE",
+              "type": "EEF",
+              "format": "XYZ_ROT6D",
+              "state_key": "right_wrist_eef"
+            },
+            {
+              "rep": "ABSOLUTE",
+              "type": "NON_EEF",
+              "format": "DEFAULT",
+              "state_key": "left_gripper_pos"
+            },
+            {
+              "rep": "ABSOLUTE",
+              "type": "NON_EEF",
+              "format": "DEFAULT",
+              "state_key": "right_gripper_pos"
+            },
+            {
+              "rep": "RELATIVE",
+              "type": "NON_EEF",
+              "format": "DEFAULT",
+              "state_key": "left_joint_pos"
+            },
+            {
+              "rep": "RELATIVE",
+              "type": "NON_EEF",
+              "format": "DEFAULT",
+              "state_key": "right_joint_pos"
+            }
+          ]
+        },
+        "language": {
+          "delta_indices": [
+            0
+          ],
+          "modality_keys": [
+            "annotation.task"
+          ]
+        }
+      },
+      "real_r1_pro_sharpa_relative_eef_maxinsights": {
+        "video": {
+          "delta_indices": [
+            -30,
+            0
+          ],
+          "modality_keys": [
+            "ego_view_cropratio_res320x240_freq30"
+          ]
+        },
+        "state": {
+          "delta_indices": [
+            0
+          ],
+          "modality_keys": [
+            "left_wrist_eef",
+            "right_wrist_eef",
+            "left_hand_joints",
+            "right_hand_joints"
+          ]
+        },
+        "action": {
+          "delta_indices": [
+            0,
+            1,
+            2,
+            3,
+            4,
+            5,
+            6,
+            7,
+            8,
+            9,
+            10,
+            11,
+            12,
+            13,
+            14,
+            15,
+            16,
+            17,
+            18,
+            19,
+            20,
+            21,
+            22,
+            23,
+            24,
+            25,
+            26,
+            27,
+            28,
+            29,
+            30,
+            31,
+            32,
+            33,
+            34,
+            35,
+            36,
+            37,
+            38,
+            39
+          ],
+          "modality_keys": [
+            "left_wrist_eef",
+            "right_wrist_eef",
+            "left_hand_joints",
+            "right_hand_joints"
+          ],
+          "action_configs": [
+            {
+              "rep": "RELATIVE",
+              "type": "EEF",
+              "format": "XYZ_ROT6D",
+              "state_key": "left_wrist_eef"
+            },
+            {
+              "rep": "RELATIVE",
+              "type": "EEF",
+              "format": "XYZ_ROT6D",
+              "state_key": "right_wrist_eef"
+            },
+            {
+              "rep": "ABSOLUTE",
+              "type": "NON_EEF",
+              "format": "DEFAULT",
+              "state_key": "left_hand_joints"
+            },
+            {
+              "rep": "ABSOLUTE",
+              "type": "NON_EEF",
+              "format": "DEFAULT",
+              "state_key": "right_hand_joints"
+            }
+          ]
+        },
+        "language": {
+          "delta_indices": [
+            0
+          ],
+          "modality_keys": [
+            "annotation.human.coarse_action"
+          ]
+        }
+      },
+      "xdof_relative_eef_relative_joint_subtask": {
+        "video": {
+          "delta_indices": [
+            -30,
+            0
+          ],
+          "modality_keys": [
+            "top_camera-images-rgb_320_240",
+            "left_camera-images-rgb_320_240",
+            "right_camera-images-rgb_320_240"
+          ]
+        },
+        "state": {
+          "delta_indices": [
+            0
+          ],
+          "modality_keys": [
+            "left_wrist_eef",
+            "right_wrist_eef",
+            "left_gripper_pos",
+            "right_gripper_pos",
+            "left_joint_pos",
+            "right_joint_pos"
+          ]
+        },
+        "action": {
+          "delta_indices": [
+            0,
+            1,
+            2,
+            3,
+            4,
+            5,
+            6,
+            7,
+            8,
+            9,
+            10,
+            11,
+            12,
+            13,
+            14,
+            15,
+            16,
+            17,
+            18,
+            19,
+            20,
+            21,
+            22,
+            23,
+            24,
+            25,
+            26,
+            27,
+            28,
+            29,
+            30,
+            31,
+            32,
+            33,
+            34,
+            35,
+            36,
+            37,
+            38,
+            39
+          ],
+          "modality_keys": [
+            "left_wrist_eef",
+            "right_wrist_eef",
+            "left_gripper_pos",
+            "right_gripper_pos",
+            "left_joint_pos",
+            "right_joint_pos"
+          ],
+          "action_configs": [
+            {
+              "rep": "RELATIVE",
+              "type": "EEF",
+              "format": "XYZ_ROT6D",
+              "state_key": "left_wrist_eef"
+            },
+            {
+              "rep": "RELATIVE",
+              "type": "EEF",
+              "format": "XYZ_ROT6D",
+              "state_key": "right_wrist_eef"
+            },
+            {
+              "rep": "ABSOLUTE",
+              "type": "NON_EEF",
+              "format": "DEFAULT",
+              "state_key": "left_gripper_pos"
+            },
+            {
+              "rep": "ABSOLUTE",
+              "type": "NON_EEF",
+              "format": "DEFAULT",
+              "state_key": "right_gripper_pos"
+            },
+            {
+              "rep": "RELATIVE",
+              "type": "NON_EEF",
+              "format": "DEFAULT",
+              "state_key": "left_joint_pos"
+            },
+            {
+              "rep": "RELATIVE",
+              "type": "NON_EEF",
+              "format": "DEFAULT",
+              "state_key": "right_joint_pos"
+            }
+          ]
+        },
+        "language": {
+          "delta_indices": [
+            0
+          ],
+          "modality_keys": [
+            "annotation.sub_task"
+          ]
+        }
+      },
+      "oxe_droid_relative_eef_relative_joint": {
+        "video": {
+          "delta_indices": [
+            -15,
+            0
+          ],
+          "modality_keys": [
+            "exterior_image_1_left",
+            "wrist_image_left"
+          ]
+        },
+        "state": {
+          "delta_indices": [
+            0
+          ],
+          "modality_keys": [
+            "eef_9d",
+            "gripper_position",
+            "joint_position"
+          ]
+        },
+        "action": {
+          "delta_indices": [
+            0,
+            1,
+            2,
+            3,
+            4,
+            5,
+            6,
+            7,
+            8,
+            9,
+            10,
+            11,
+            12,
+            13,
+            14,
+            15,
+            16,
+            17,
+            18,
+            19,
+            20,
+            21,
+            22,
+            23,
+            24,
+            25,
+            26,
+            27,
+            28,
+            29,
+            30,
+            31,
+            32,
+            33,
+            34,
+            35,
+            36,
+            37,
+            38,
+            39
+          ],
+          "modality_keys": [
+            "eef_9d",
+            "gripper_position",
+            "joint_position"
+          ],
+          "action_configs": [
+            {
+              "rep": "RELATIVE",
+              "type": "EEF",
+              "format": "XYZ_ROT6D",
+              "state_key": "eef_9d"
+            },
+            {
+              "rep": "ABSOLUTE",
+              "type": "NON_EEF",
+              "format": "DEFAULT",
+              "state_key": "gripper_position"
+            },
+            {
+              "rep": "RELATIVE",
+              "type": "NON_EEF",
+              "format": "DEFAULT",
+              "state_key": "joint_position"
+            }
+          ]
+        },
+        "language": {
+          "delta_indices": [
+            0
+          ],
+          "modality_keys": [
+            "annotation.language.language_instruction"
+          ]
+        }
+      }
+    },
+    "use_percentiles": true,
+    "use_mean_std": false,
+    "image_crop_size": [
+      230,
+      230
+    ],
+    "image_target_size": [
+      256,
+      256
+    ],
+    "formalize_language": true,
+    "max_state_dim": 132,
+    "max_action_dim": 132,
+    "apply_sincos_state_encoding": false,
+    "color_jitter_params": {
+      "brightness": 0.3,
+      "contrast": 0.4,
+      "saturation": 0.5,
+      "hue": 0.08
+    },
+    "random_rotation_angle": 0,
+    "letter_box_transform": false,
+    "exclude_state": false,
+    "state_dropout_prob": 0.2,
+    "use_albumentations": true,
+    "shortest_image_edge": 256,
+    "crop_fraction": 0.95,
+    "max_action_horizon": 40,
+    "use_relative_action": true
+  }
+}

scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5ae78127cbbf7977166b80c05e2b2a0f95f6a251fd8b8fbac5f86c0cfe018ffe
+size 1263

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3c45bcc6f4918ba6ffc8c1dd3aca5d223778a483fcede2c19948cc43f206fc0d
+size 8259

wandb_config.json

@@ -0,0 +1 @@
+{"project": "human_pretraining_n15_galaxea_sharpa", "run_id": "pretrain_n17_qwen3vl_2b_finetuned_sft_albumentations_2frame_abshand_frozen_vlm_double_dit_vl_interleaved_lr5e-5_200ksteps_batchsize8192"}

zero_to_fp32.py

@@ -0,0 +1,760 @@
+#!/usr/bin/env python
+
+# Copyright (c) Microsoft Corporation.
+# SPDX-License-Identifier: Apache-2.0
+
+# DeepSpeed Team
+
+# This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets
+# copied into the top level checkpoint dir, so the user can easily do the conversion at any point in
+# the future. Once extracted, the weights don't require DeepSpeed and can be used in any
+# application.
+#
+# example:
+#   python zero_to_fp32.py . output_dir/
+#   or
+#   python zero_to_fp32.py . output_dir/ --safe_serialization
+
+import argparse
+import torch
+import glob
+import math
+import os
+import re
+import gc
+import json
+import numpy as np
+from tqdm import tqdm
+from collections import OrderedDict
+from dataclasses import dataclass
+
+# while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with
+# DeepSpeed data structures it has to be available in the current python environment.
+from deepspeed.utils import logger
+from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS,
+                                            FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES,
+                                            FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS)
+
+
+@dataclass
+class zero_model_state:
+    buffers: dict()
+    param_shapes: dict()
+    shared_params: list
+    ds_version: int
+    frozen_param_shapes: dict()
+    frozen_param_fragments: dict()
+
+
+debug = 0
+
+# load to cpu
+device = torch.device('cpu')
+
+
+def atoi(text):
+    return int(text) if text.isdigit() else text
+
+
+def natural_keys(text):
+    '''
+    alist.sort(key=natural_keys) sorts in human order
+    http://nedbatchelder.com/blog/200712/human_sorting.html
+    (See Toothy's implementation in the comments)
+    '''
+    return [atoi(c) for c in re.split(r'(\d+)', text)]
+
+
+def get_model_state_file(checkpoint_dir, zero_stage):
+    if not os.path.isdir(checkpoint_dir):
+        raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist")
+
+    # there should be only one file
+    if zero_stage <= 2:
+        file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt")
+    elif zero_stage == 3:
+        file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt")
+
+    if not os.path.exists(file):
+        raise FileNotFoundError(f"can't find model states file at '{file}'")
+
+    return file
+
+
+def get_checkpoint_files(checkpoint_dir, glob_pattern):
+    # XXX: need to test that this simple glob rule works for multi-node setup too
+    ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys)
+
+    if len(ckpt_files) == 0:
+        raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'")
+
+    return ckpt_files
+
+
+def get_optim_files(checkpoint_dir):
+    return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt")
+
+
+def get_model_state_files(checkpoint_dir):
+    return get_checkpoint_files(checkpoint_dir, "*_model_states.pt")
+
+
+def parse_model_states(files):
+    zero_model_states = []
+    for file in files:
+        state_dict = torch.load(file, map_location=device, weights_only=False)
+
+        if BUFFER_NAMES not in state_dict:
+            raise ValueError(f"{file} is not a model state checkpoint")
+        buffer_names = state_dict[BUFFER_NAMES]
+        if debug:
+            print("Found buffers:", buffer_names)
+
+        # recover just the buffers while restoring them to fp32 if they were saved in fp16
+        buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names}
+        param_shapes = state_dict[PARAM_SHAPES]
+
+        # collect parameters that are included in param_shapes
+        param_names = []
+        for s in param_shapes:
+            for name in s.keys():
+                param_names.append(name)
+
+        # update with frozen parameters
+        frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None)
+        if frozen_param_shapes is not None:
+            if debug:
+                print(f"Found frozen_param_shapes: {frozen_param_shapes}")
+            param_names += list(frozen_param_shapes.keys())
+
+        # handle shared params
+        shared_params = [[k, v] for k, v in state_dict["shared_params"].items()]
+
+        ds_version = state_dict.get(DS_VERSION, None)
+
+        frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None)
+
+        z_model_state = zero_model_state(buffers=buffers,
+                                         param_shapes=param_shapes,
+                                         shared_params=shared_params,
+                                         ds_version=ds_version,
+                                         frozen_param_shapes=frozen_param_shapes,
+                                         frozen_param_fragments=frozen_param_fragments)
+        zero_model_states.append(z_model_state)
+
+    return zero_model_states
+
+
+def parse_optim_states(files, ds_checkpoint_dir):
+    total_files = len(files)
+    state_dicts = []
+    for f in tqdm(files, desc='Loading checkpoint shards'):
+        state_dict = torch.load(f, map_location=device, mmap=True, weights_only=False)
+        # immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights
+        # and also handle the case where it was already removed by another helper script
+        state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None)
+        state_dicts.append(state_dict)
+
+    if ZERO_STAGE not in state_dicts[0][OPTIMIZER_STATE_DICT]:
+        raise ValueError(f"{files[0]} is not a zero checkpoint")
+    zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE]
+    world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT]
+
+    # For ZeRO-2 each param group can have different partition_count as data parallelism for expert
+    # parameters can be different from data parallelism for non-expert parameters. So we can just
+    # use the max of the partition_count to get the dp world_size.
+
+    if type(world_size) is list:
+        world_size = max(world_size)
+
+    if world_size != total_files:
+        raise ValueError(
+            f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. "
+            "Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes."
+        )
+
+    # the groups are named differently in each stage
+    if zero_stage <= 2:
+        fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS
+    elif zero_stage == 3:
+        fp32_groups_key = FP32_FLAT_GROUPS
+    else:
+        raise ValueError(f"unknown zero stage {zero_stage}")
+
+    fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))]
+    return zero_stage, world_size, fp32_flat_groups
+
+
+def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters):
+    """
+    Returns fp32 state_dict reconstructed from ds checkpoint
+
+    Args:
+        - ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are)
+
+    """
+    print(f"Processing zero checkpoint '{ds_checkpoint_dir}'")
+
+    optim_files = get_optim_files(ds_checkpoint_dir)
+    zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir)
+    print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}")
+
+    model_files = get_model_state_files(ds_checkpoint_dir)
+
+    zero_model_states = parse_model_states(model_files)
+    print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}')
+
+    if zero_stage <= 2:
+        return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                                          exclude_frozen_parameters)
+    elif zero_stage == 3:
+        return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                                          exclude_frozen_parameters)
+
+
+def _zero2_merge_frozen_params(state_dict, zero_model_states):
+    if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0:
+        return
+
+    frozen_param_shapes = zero_model_states[0].frozen_param_shapes
+    frozen_param_fragments = zero_model_states[0].frozen_param_fragments
+
+    if debug:
+        num_elem = sum(s.numel() for s in frozen_param_shapes.values())
+        print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}')
+
+        wanted_params = len(frozen_param_shapes)
+        wanted_numel = sum(s.numel() for s in frozen_param_shapes.values())
+        avail_numel = sum([p.numel() for p in frozen_param_fragments.values()])
+        print(f'Frozen params: Have {avail_numel} numels to process.')
+        print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params')
+
+    total_params = 0
+    total_numel = 0
+    for name, shape in frozen_param_shapes.items():
+        total_params += 1
+        unpartitioned_numel = shape.numel()
+        total_numel += unpartitioned_numel
+
+        state_dict[name] = frozen_param_fragments[name]
+
+        if debug:
+            print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ")
+
+    print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _has_callable(obj, fn):
+    attr = getattr(obj, fn, None)
+    return callable(attr)
+
+
+def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states):
+    param_shapes = zero_model_states[0].param_shapes
+
+    # Reconstruction protocol:
+    #
+    # XXX: document this
+
+    if debug:
+        for i in range(world_size):
+            for j in range(len(fp32_flat_groups[0])):
+                print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}")
+
+    # XXX: memory usage doubles here (zero2)
+    num_param_groups = len(fp32_flat_groups[0])
+    merged_single_partition_of_fp32_groups = []
+    for i in range(num_param_groups):
+        merged_partitions = [sd[i] for sd in fp32_flat_groups]
+        full_single_fp32_vector = torch.cat(merged_partitions, 0)
+        merged_single_partition_of_fp32_groups.append(full_single_fp32_vector)
+    avail_numel = sum(
+        [full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups])
+
+    if debug:
+        wanted_params = sum([len(shapes) for shapes in param_shapes])
+        wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes])
+        # not asserting if there is a mismatch due to possible padding
+        print(f"Have {avail_numel} numels to process.")
+        print(f"Need {wanted_numel} numels in {wanted_params} params.")
+
+    # params
+    # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support
+    # out-of-core computing solution
+    total_numel = 0
+    total_params = 0
+    for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups):
+        offset = 0
+        avail_numel = full_single_fp32_vector.numel()
+        for name, shape in shapes.items():
+
+            unpartitioned_numel = shape.numel() if _has_callable(shape, 'numel') else math.prod(shape)
+            total_numel += unpartitioned_numel
+            total_params += 1
+
+            if debug:
+                print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ")
+            state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape)
+            offset += unpartitioned_numel
+
+        # Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and
+        # avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex
+        # paddings performed in the code it's almost impossible to predict the exact numbers w/o the
+        # live optimizer object, so we are checking that the numbers are within the right range
+        align_to = 2 * world_size
+
+        def zero2_align(x):
+            return align_to * math.ceil(x / align_to)
+
+        if debug:
+            print(f"original offset={offset}, avail_numel={avail_numel}")
+
+        offset = zero2_align(offset)
+        avail_numel = zero2_align(avail_numel)
+
+        if debug:
+            print(f"aligned  offset={offset}, avail_numel={avail_numel}")
+
+        # Sanity check
+        if offset != avail_numel:
+            raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong")
+
+    print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                               exclude_frozen_parameters):
+    state_dict = OrderedDict()
+
+    # buffers
+    buffers = zero_model_states[0].buffers
+    state_dict.update(buffers)
+    if debug:
+        print(f"added {len(buffers)} buffers")
+
+    if not exclude_frozen_parameters:
+        _zero2_merge_frozen_params(state_dict, zero_model_states)
+
+    _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states)
+
+    # recover shared parameters
+    for pair in zero_model_states[0].shared_params:
+        if pair[1] in state_dict:
+            state_dict[pair[0]] = state_dict[pair[1]]
+
+    return state_dict
+
+
+def zero3_partitioned_param_info(unpartitioned_numel, world_size):
+    remainder = unpartitioned_numel % world_size
+    padding_numel = (world_size - remainder) if remainder else 0
+    partitioned_numel = math.ceil(unpartitioned_numel / world_size)
+    return partitioned_numel, padding_numel
+
+
+def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states):
+    if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0:
+        return
+
+    if debug:
+        for i in range(world_size):
+            num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values())
+            print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}')
+
+        frozen_param_shapes = zero_model_states[0].frozen_param_shapes
+        wanted_params = len(frozen_param_shapes)
+        wanted_numel = sum(s.numel() for s in frozen_param_shapes.values())
+        avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size
+        print(f'Frozen params: Have {avail_numel} numels to process.')
+        print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params')
+
+    total_params = 0
+    total_numel = 0
+    for name, shape in zero_model_states[0].frozen_param_shapes.items():
+        total_params += 1
+        unpartitioned_numel = shape.numel()
+        total_numel += unpartitioned_numel
+
+        param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states)
+        state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape)
+
+        partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size)
+
+        if debug:
+            print(
+                f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}"
+            )
+
+    print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements")
+
+
+class GatheredTensor:
+    """
+    A pseudo tensor that collects partitioned weights.
+    It is more memory efficient when there are multiple groups.
+    """
+
+    def __init__(self, flat_groups, flat_groups_offset, offset, partitioned_numel, shape):
+        self.flat_groups = flat_groups
+        self.flat_groups_offset = flat_groups_offset
+        self.offset = offset
+        self.partitioned_numel = partitioned_numel
+        self.shape = shape
+        self.dtype = self.flat_groups[0][0].dtype
+
+    def contiguous(self):
+        """
+        Merge partitioned weights from flat_groups into a single tensor.
+        """
+        end_idx = self.offset + self.partitioned_numel
+        world_size = len(self.flat_groups)
+        pad_flat_param_chunks = []
+
+        for rank_i in range(world_size):
+            # for each rank, we need to collect weights from related group/groups
+            flat_groups_at_rank_i = self.flat_groups[rank_i]
+            start_group_id = None
+            end_group_id = None
+            for group_id in range(len(self.flat_groups_offset)):
+                if self.flat_groups_offset[group_id] <= self.offset < self.flat_groups_offset[group_id + 1]:
+                    start_group_id = group_id
+                if self.flat_groups_offset[group_id] < end_idx <= self.flat_groups_offset[group_id + 1]:
+                    end_group_id = group_id
+                    break
+            # collect weights from related group/groups
+            for group_id in range(start_group_id, end_group_id + 1):
+                flat_tensor = flat_groups_at_rank_i[group_id]
+                start_offset = self.offset - self.flat_groups_offset[group_id]
+                end_offset = min(end_idx, self.flat_groups_offset[group_id + 1]) - self.flat_groups_offset[group_id]
+                pad_flat_param_chunks.append(flat_tensor[start_offset:end_offset])
+
+        # collect weights from all ranks
+        pad_flat_param = torch.cat(pad_flat_param_chunks, dim=0)
+        param = pad_flat_param[:self.shape.numel()].view(self.shape).contiguous()
+        return param
+
+
+def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states):
+    param_shapes = zero_model_states[0].param_shapes
+    avail_numel = sum([flat_group.numel() for flat_group in fp32_flat_groups[0]]) * world_size
+
+    # Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each
+    # param, re-consolidating each param, while dealing with padding if any
+
+    # merge list of dicts, preserving order
+    param_shapes = {k: v for d in param_shapes for k, v in d.items()}
+
+    if debug:
+        for i in range(world_size):
+            print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}")
+
+        wanted_params = len(param_shapes)
+        wanted_numel = sum(shape.numel() for shape in param_shapes.values())
+        # not asserting if there is a mismatch due to possible padding
+        avail_numel = fp32_flat_groups[0].numel() * world_size
+        print(f"Trainable params: Have {avail_numel} numels to process.")
+        print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.")
+
+    # params
+    # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support
+    # out-of-core computing solution
+    offset = 0
+    total_numel = 0
+    total_params = 0
+    flat_groups_offset = [0] + list(np.cumsum([flat_tensor.numel() for flat_tensor in fp32_flat_groups[0]]))
+    for name, shape in tqdm(param_shapes.items(), desc='Gathering sharded weights'):
+        unpartitioned_numel = shape.numel()
+        total_numel += unpartitioned_numel
+        total_params += 1
+        partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size)
+
+        if debug:
+            print(
+                f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}"
+            )
+
+        # memory efficient tensor
+        tensor = GatheredTensor(fp32_flat_groups, flat_groups_offset, offset, partitioned_numel, shape)
+        state_dict[name] = tensor
+        offset += partitioned_numel
+
+    offset *= world_size
+
+    # Sanity check
+    if offset != avail_numel:
+        raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong")
+
+    print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                               exclude_frozen_parameters):
+    state_dict = OrderedDict()
+
+    # buffers
+    buffers = zero_model_states[0].buffers
+    state_dict.update(buffers)
+    if debug:
+        print(f"added {len(buffers)} buffers")
+
+    if not exclude_frozen_parameters:
+        _zero3_merge_frozen_params(state_dict, world_size, zero_model_states)
+
+    _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states)
+
+    # recover shared parameters
+    for pair in zero_model_states[0].shared_params:
+        if pair[1] in state_dict:
+            state_dict[pair[0]] = state_dict[pair[1]]
+
+    return state_dict
+
+
+def to_torch_tensor(state_dict, return_empty_tensor=False):
+    """
+    Convert state_dict of GatheredTensor to torch tensor
+    """
+    torch_state_dict = {}
+    converted_tensors = {}
+    for name, tensor in state_dict.items():
+        tensor_id = id(tensor)
+        if tensor_id in converted_tensors:  # shared tensors
+            shared_tensor = torch_state_dict[converted_tensors[tensor_id]]
+            torch_state_dict[name] = shared_tensor
+        else:
+            converted_tensors[tensor_id] = name
+            if return_empty_tensor:
+                torch_state_dict[name] = torch.empty(tensor.shape, dtype=tensor.dtype)
+            else:
+                torch_state_dict[name] = tensor.contiguous()
+    return torch_state_dict
+
+
+def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir,
+                                             tag=None,
+                                             exclude_frozen_parameters=False,
+                                             lazy_mode=False):
+    """
+    Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with
+    ``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example
+    via a model hub.
+
+    Args:
+        - ``checkpoint_dir``: path to the desired checkpoint folder
+        - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14``
+        - ``exclude_frozen_parameters``: exclude frozen parameters
+        - ``lazy_mode``: get state_dict in lazy mode. It returns a dict of pesduo tensor instead of torch tensor, which is more memory efficient.
+          Convert the pesduo tensor to torch tensor by ``.contiguous()``
+
+    Returns:
+        - pytorch ``state_dict``
+
+    A typical usage might be ::
+
+        from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint
+        # do the training and checkpoint saving
+        state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu
+        model = model.cpu() # move to cpu
+        model.load_state_dict(state_dict)
+        # submit to model hub or save the model to share with others
+
+    In this example the ``model`` will no longer be usable in the deepspeed context of the same
+    application. i.e. you will need to re-initialize the deepspeed engine, since
+    ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it.
+
+    If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead.
+
+    Note: the above usage may not work if your application doesn't have sufficient free CPU memory.
+    You may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with
+    the checkpoint. Or you can load state_dict in lazy mode ::
+
+        from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint
+        state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, lazy_mode=True) # not on cpu
+        for name, lazy_tensor in state_dict.item():
+            tensor = lazy_tensor.contiguous()  # to cpu
+            print(name, tensor)
+            # del tensor to release memory if it no longer in use
+    """
+    if tag is None:
+        latest_path = os.path.join(checkpoint_dir, 'latest')
+        if os.path.isfile(latest_path):
+            with open(latest_path, 'r') as fd:
+                tag = fd.read().strip()
+        else:
+            raise ValueError(f"Unable to find 'latest' file at {latest_path}")
+
+    ds_checkpoint_dir = os.path.join(checkpoint_dir, tag)
+
+    if not os.path.isdir(ds_checkpoint_dir):
+        raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist")
+
+    state_dict = _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters)
+    if lazy_mode:
+        return state_dict
+    else:
+        return to_torch_tensor(state_dict)
+
+
+def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir,
+                                               output_dir,
+                                               max_shard_size="5GB",
+                                               safe_serialization=False,
+                                               tag=None,
+                                               exclude_frozen_parameters=False):
+    """
+    Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be
+    loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed.
+
+    Args:
+        - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``)
+        - ``output_dir``: directory to the pytorch fp32 state_dict output files
+        - ``max_shard_size``: the maximum size for a checkpoint before being sharded, default value is 5GB
+        - ``safe_serialization``:  whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`).
+        - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14``
+        - ``exclude_frozen_parameters``: exclude frozen parameters
+    """
+
+    # Dependency pre-check
+    if safe_serialization:
+        try:
+            from safetensors.torch import save_file
+        except ImportError:
+            print('If you want to use `safe_serialization`, please `pip install safetensors`')
+            raise
+    if max_shard_size is not None:
+        try:
+            from huggingface_hub import split_torch_state_dict_into_shards
+        except ImportError:
+            print('If you want to use `max_shard_size`, please `pip install huggingface_hub`')
+            raise
+
+    # Convert zero checkpoint to state_dict
+    state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir,
+                                                          tag,
+                                                          exclude_frozen_parameters,
+                                                          lazy_mode=True)
+
+    # Shard the model if it is too big.
+    weights_name = "model.safetensors" if safe_serialization else "pytorch_model.bin"
+    if max_shard_size is not None:
+        filename_pattern = weights_name.replace(".bin", "{suffix}.bin").replace(".safetensors", "{suffix}.safetensors")
+        # an memory-efficient approach for sharding
+        empty_state_dict = to_torch_tensor(state_dict, return_empty_tensor=True)
+        state_dict_split = split_torch_state_dict_into_shards(empty_state_dict,
+                                                              filename_pattern=filename_pattern,
+                                                              max_shard_size=max_shard_size)
+    else:
+        from collections import namedtuple
+        StateDictSplit = namedtuple("StateDictSplit", ["is_sharded", "filename_to_tensors"])
+        state_dict_split = StateDictSplit(is_sharded=False,
+                                          filename_to_tensors={weights_name: list(state_dict.keys())})
+
+    # Save the model by shard
+    os.makedirs(output_dir, exist_ok=True)
+    filename_to_tensors = state_dict_split.filename_to_tensors.items()
+    for shard_file, tensors in tqdm(filename_to_tensors, desc="Saving checkpoint shards"):
+        shard_state_dict = {tensor_name: state_dict[tensor_name] for tensor_name in tensors}
+        shard_state_dict = to_torch_tensor(shard_state_dict)
+        output_path = os.path.join(output_dir, shard_file)
+        if safe_serialization:
+            save_file(shard_state_dict, output_path, metadata={"format": "pt"})
+        else:
+            torch.save(shard_state_dict, output_path)
+        # release the memory of current shard
+        for tensor_name in list(shard_state_dict.keys()):
+            del state_dict[tensor_name]
+            del shard_state_dict[tensor_name]
+        del shard_state_dict
+        gc.collect()
+
+    # Save index if sharded
+    if state_dict_split.is_sharded:
+        index = {
+            "metadata": state_dict_split.metadata,
+            "weight_map": state_dict_split.tensor_to_filename,
+        }
+        save_index_file = "model.safetensors.index.json" if safe_serialization else "pytorch_model.bin.index.json"
+        save_index_file = os.path.join(output_dir, save_index_file)
+        with open(save_index_file, "w", encoding="utf-8") as f:
+            content = json.dumps(index, indent=2, sort_keys=True) + "\n"
+            f.write(content)
+
+
+def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None):
+    """
+    1. Put the provided model to cpu
+    2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict``
+    3. Load it into the provided model
+
+    Args:
+        - ``model``: the model object to update
+        - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``)
+        - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14``
+
+    Returns:
+        - ``model`: modified model
+
+    Make sure you have plenty of CPU memory available before you call this function. If you don't
+    have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it
+    conveniently placed for you in the checkpoint folder.
+
+    A typical usage might be ::
+
+        from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint
+        model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir)
+        # submit to model hub or save the model to share with others
+
+    Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context
+    of the same application. i.e. you will need to re-initialize the deepspeed engine, since
+    ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it.
+
+    """
+    logger.info("Extracting fp32 weights")
+    state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag)
+
+    logger.info("Overwriting model with fp32 weights")
+    model = model.cpu()
+    model.load_state_dict(state_dict, strict=False)
+
+    return model
+
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser()
+    parser.add_argument("checkpoint_dir",
+                        type=str,
+                        help="path to the desired checkpoint folder, e.g., path/checkpoint-12")
+    parser.add_argument("output_dir",
+                        type=str,
+                        help="directory to the pytorch fp32 state_dict output files"
+                        "(e.g. path/checkpoint-12-output/)")
+    parser.add_argument(
+        "--max_shard_size",
+        type=str,
+        default="5GB",
+        help="The maximum size for a checkpoint before being sharded. Checkpoints shard will then be each of size"
+        "lower than this size. If expressed as a string, needs to be digits followed by a unit (like `5MB`"
+        "We default it to 5GB in order for models to be able to run easily on free-tier google colab instances"
+        "without CPU OOM issues.")
+    parser.add_argument(
+        "--safe_serialization",
+        default=False,
+        action='store_true',
+        help="Whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`).")
+    parser.add_argument("-t",
+                        "--tag",
+                        type=str,
+                        default=None,
+                        help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1")
+    parser.add_argument("--exclude_frozen_parameters", action='store_true', help="exclude frozen parameters")
+    parser.add_argument("-d", "--debug", action='store_true', help="enable debug")
+    args = parser.parse_args()
+
+    debug = args.debug
+
+    convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir,
+                                               args.output_dir,
+                                               max_shard_size=args.max_shard_size,
+                                               safe_serialization=args.safe_serialization,
+                                               tag=args.tag,
+                                               exclude_frozen_parameters=args.exclude_frozen_parameters)