Upload PyTorch model

README.md

@@ -1,245 +1,75 @@
 ---
-license: other
-base_model: DedalusHealthCare/tinybert-mlm-de
-datasets:
-- DedalusHealthCare/ner_demo_de
-task_categories:
-- token-classification
-task_ids:
-- named-entity-recognition
+library_name: transformers
 language:
 - de
+license: other
+base_model: DedalusHealthCare/tinybert-mlm-de
 tags:
-- token-classification
-- ner
-- named-entity-recognition
-- de
-- disorder_finding
-library_name: transformers
-pipeline_tag: token-classification
+- generated_from_trainer
+datasets:
+- demo-de
+model-index:
+- name: tinybert-clinalytix_1774355441
+  results: []
 ---
 
-# TinyBERT for Demo NER (German)
-
-## Model Description
-
-This model is a fine-tuned TinyBERT model for Named Entity Recognition (NER) of DISORDER_FINDING entities in German medical texts.
-
-It was fine-tuned from the [DedalusHealthCare/tinybert-mlm-de](https://huggingface.co/DedalusHealthCare/tinybert-mlm-de) masked language model using the [DedalusHealthCare/ner_demo_de](https://huggingface.co/datasets/DedalusHealthCare/ner_demo_de) dataset.
-
-**Base Model**: [DedalusHealthCare/tinybert-mlm-de](https://huggingface.co/DedalusHealthCare/tinybert-mlm-de)
-
-**Training Dataset**: [DedalusHealthCare/ner_demo_de](https://huggingface.co/datasets/DedalusHealthCare/ner_demo_de)
-
-**Task**: Token Classification (Named Entity Recognition)
-
-**Language**: German (de)
-
-**Entities**: DISORDER_FINDING
-
-**Model Format**: PYTORCH+ONNX
-
-**Please use `max` as aggregation strategy in the NER pipeline (see example below)**.
-
-## Training Details
-
-- **Training epochs**: 1
-- **Learning rate**: N/A
-- **Training batch size**: 32
-- **Evaluation batch size**: 32
-- **Max sequence length**: 256
-- **Warmup steps**: N/A
-- **FP16**: False
-- **Gradient accumulation steps**: 2
-- **Evaluation accumulation steps**: 2
-- **Save steps**: 15000
-- **Evaluation steps**: 10000
-- **Evaluation strategy**: steps
-- **Random seed**: 33
-- **Label all tokens**: True
-- **Balanced training**: False
-- **Chunk mode**: sliding_window
-- **Stride**: 16
-- **Max training samples**: None
-- **Max evaluation samples**: 10000
-- **Early stopping patience**: 0
-- **Early stopping threshold**: 0.0
-
-## Use Case Configuration
-
-- **Use case name**: demo
-- **Language**: German (de)
-- **Target entities**: DISORDER_FINDING
-- **Text processing max length**: N/A
-- **Entity labeling scheme**: N/A
-
-## Usage
-
-### Using Transformers Pipeline
-
-```python
-from transformers import pipeline
-
-# Load the model
-ner_pipeline = pipeline(
-    "ner",
-    model="DedalusHealthCare/tinybert-ner-demo-de",
-    tokenizer="DedalusHealthCare/tinybert-ner-demo-de",
-    aggregation_strategy="max"
-)
-
-# Example text
-text = "Der Patient hat Diabetes und Bluthochdruck."
-
-# Get predictions
-entities = ner_pipeline(text)
-print(entities)
-```
-
-### Using AutoModel and AutoTokenizer
-
-```python
-from transformers import AutoTokenizer, AutoModelForTokenClassification
-import torch
-
-# Load model and tokenizer
-model_name = "DedalusHealthCare/tinybert-ner-demo-de"
-tokenizer = AutoTokenizer.from_pretrained(model_name)
-model = AutoModelForTokenClassification.from_pretrained(model_name)
-
-# Tokenize text
-text = "Der Patient hat Diabetes und Bluthochdruck."
-tokens = tokenizer(text, return_tensors="pt", truncation=True, padding=True)
-
-# Get predictions
-with torch.no_grad():
-    outputs = model(**tokens)
-    predictions = torch.nn.functional.softmax(outputs.logits, dim=-1)
-
-# Get labels
-predicted_token_class_ids = predictions.argmax(-1)
-labels = [model.config.id2label[id.item()] for id in predicted_token_class_ids[0]]
-```
-
-### Using ONNX Runtime (Optimized Inference)
-
-```python
-from optimum.onnxruntime import ORTModelForTokenClassification
-from transformers import AutoTokenizer, pipeline
-import torch
-
-# Load ONNX model for faster inference
-model_name = "DedalusHealthCare/tinybert-ner-demo-de"
-onnx_model = ORTModelForTokenClassification.from_pretrained(model_name)
-tokenizer = AutoTokenizer.from_pretrained(model_name)
-
-# Create pipeline with ONNX model (recommended)
-ner_pipeline = pipeline(
-    "ner",
-    model=onnx_model,
-    tokenizer=tokenizer,
-    aggregation_strategy="max"
-)
-
-# Example text
-text = "Der Patient hat Diabetes und Bluthochdruck."
-entities = ner_pipeline(text)
-print(entities)
-
-# Direct model usage
-inputs = tokenizer(text, return_tensors="pt", truncation=True, padding=True)
-with torch.no_grad():
-    outputs = onnx_model(**inputs)
-    predictions = torch.nn.functional.softmax(outputs.logits, dim=-1)
-
-predicted_token_class_ids = predictions.argmax(-1)
-token_labels = [onnx_model.config.id2label[id.item()] for id in predicted_token_class_ids[0]]
-```
-
-### Performance Comparison
-
-- **PyTorch**: Standard format, suitable for training and research
-- **ONNX**: Optimized for inference, typically 2-4x faster than PyTorch
-- **Recommendation**: Use ONNX for production inference, PyTorch for research
-
-## Model Architecture
-
-This model is based on the TinyBERT architecture with a token classification head for Named Entity Recognition.
-
-## Intended Use
-
-This model is intended for:
-- Named Entity Recognition in German medical texts
-- Identification of DISORDER_FINDING entities
-- Medical text processing and analysis
-- Research and development in medical NLP
-
-## Limitations
-
-- Trained specifically for German medical texts
-- Performance may vary on texts from different medical domains
-- May not generalize well to non-medical texts
-- Requires careful evaluation on new datasets
-
-## Ethical Considerations
-
-- This model is trained on medical data and should be used responsibly
-- Outputs should be validated by medical professionals
-- Patient privacy and data protection regulations must be followed
-- The model may have biases present in the training data
-
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
 
-## Model Performance
+# tinybert-clinalytix_1774355441
 
-This model has been evaluated on the **goldset from ner_disorderfinding_de_goldset** using
-IO evaluation (sklearn, token level, lenient) with the following results:
+This model is a fine-tuned version of [DedalusHealthCare/tinybert-mlm-de](https://huggingface.co/DedalusHealthCare/tinybert-mlm-de) on the demo-de dataset.
+It achieves the following results on the evaluation set:
+- Loss: 1.7065
+- Disorder Precision: 0.0
+- Disorder Recall: 0.0
+- Disorder F1: 0.0
+- Disorder Number: 2
+- Finding Precision: 0.0
+- Finding Recall: 0.0
+- Finding F1: 0.0
+- Finding Number: 0
+- Overall Precision: 0.0
+- Overall Recall: 0.0
+- Overall F1: 0.0
+- Overall Accuracy: 0.1176
 
-### Overall Performance
+## Model description
 
-| Metric | Score |
-|--------|-------|
-| Precision (Macro) | 0.423825 |
-| Recall (Macro) | 0.467183 |
-| F1-Score (Macro) | 0.435170 |
-| Precision (Weighted) | 0.599471 |
-| Recall (Weighted) | 0.697989 |
-| F1-Score (Weighted) | 0.640426 |
+More information needed
 
-**Inference Performance**: 5.53 seconds for evaluation dataset
+## Intended uses & limitations
 
-### Entity-Level Performance (IO Evaluation)
+More information needed
 
-| Entity Type | Precision | Recall | F1-Score | Support |
-|-------------|-----------|--------|----------|---------|
-| DISORDER_FINDING | 0.753533 | 0.900434 | 0.820460 | N/A |
+## Training and evaluation data
 
-### Evaluation Details
+More information needed
 
-- **Dataset**: goldset from ner_disorderfinding_de_goldset
-- **Dataset Source**: goldset
-- **Evaluation Date**: 2025-11-03 12:25:56
-- **Language**: de
-- **Entities**: DISORDER_FINDING
+## Training procedure
 
-*This evaluation section is automatically generated and updated.*
-## Citation
+### Training hyperparameters
 
-If you use this model, please cite:
+The following hyperparameters were used during training:
+- learning_rate: 5e-05
+- train_batch_size: 32
+- eval_batch_size: 32
+- seed: 33
+- gradient_accumulation_steps: 2
+- total_train_batch_size: 64
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: linear
+- lr_scheduler_warmup_ratio: 0.1
+- num_epochs: 1
+- label_smoothing_factor: 0.1
 
-```bibtex
-@model{demo_de_ner_model,
-  title = {TinyBERT for Demo NER (German)},
-  author = {DH Healthcare GmbH},
-  year = {2025},
-  publisher = {Hugging Face},
-  url = {https://huggingface.co/DedalusHealthCare/tinybert-ner-demo-de}
-}
-```
+### Training results
 
-## License
 
-This model is proprietary and owned by DH Healthcare GmbH. All rights reserved.
 
-## Contact
+### Framework versions
 
-For questions or support, please contact DH Healthcare GmbH.
+- Transformers 4.45.1
+- Pytorch 2.10.0+cu128
+- Datasets 4.5.0
+- Tokenizers 0.20.3

all_results.json

@@ -0,0 +1,26 @@
+{
+    "epoch": 1.0,
+    "eval_DISORDER_f1": 0.0,
+    "eval_DISORDER_number": 2,
+    "eval_DISORDER_precision": 0.0,
+    "eval_DISORDER_recall": 0.0,
+    "eval_FINDING_f1": 0.0,
+    "eval_FINDING_number": 0,
+    "eval_FINDING_precision": 0.0,
+    "eval_FINDING_recall": 0.0,
+    "eval_loss": 1.7064510583877563,
+    "eval_overall_accuracy": 0.11764705882352941,
+    "eval_overall_f1": 0.0,
+    "eval_overall_precision": 0.0,
+    "eval_overall_recall": 0.0,
+    "eval_runtime": 0.1739,
+    "eval_samples": 3,
+    "eval_samples_per_second": 17.256,
+    "eval_steps_per_second": 5.752,
+    "total_flos": 21821285850.0,
+    "train_loss": 0.8611775636672974,
+    "train_runtime": 1.1366,
+    "train_samples": 17,
+    "train_samples_per_second": 14.957,
+    "train_steps_per_second": 0.88
+}

checkpoint-1/config.json

@@ -0,0 +1,42 @@
+{
+  "_name_or_path": "DedalusHealthCare/tinybert-mlm-de",
+  "architectures": [
+    "BertForTokenClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "finetuning_task": "ner",
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 312,
+  "id2label": {
+    "0": "B-DISORDER",
+    "1": "B-FINDING",
+    "2": "I-DISORDER",
+    "3": "I-FINDING",
+    "4": "O"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 312,
+  "label2id": {
+    "B-DISORDER": 0,
+    "B-FINDING": 1,
+    "I-DISORDER": 2,
+    "I-FINDING": 3,
+    "O": 4
+  },
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "bert",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 4,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "pre_trained": "",
+  "torch_dtype": "float32",
+  "training": "",
+  "transformers_version": "4.45.1",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 31102
+}

checkpoint-1/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:70bb4470bd4149e9aa0d4ab934e0e4ffc533b8af1702686320f4b3d57f8a062e
+size 48868580

checkpoint-1/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:016a14a774433cfd877144dd293cac3252b81b21b3ddd226eabd54af0507dc28
+size 97776331

checkpoint-1/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cf691b43f93b3451c34523702ad48fab1acaf3207fba0e9c194fa3aedeb5ec8c
+size 14455

checkpoint-1/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6e7a808092c9737ce7476966a087defa251dbafd354d39128871d37cbc6fa6c4
+size 1465

checkpoint-1/special_tokens_map.json

@@ -0,0 +1,37 @@
+{
+  "cls_token": {
+    "content": "[CLS]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "mask_token": {
+    "content": "[MASK]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "[PAD]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "sep_token": {
+    "content": "[SEP]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "[UNK]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

checkpoint-1/tokenizer_config.json

@@ -0,0 +1,62 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "[PAD]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "101": {
+      "content": "[UNK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "102": {
+      "content": "[CLS]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "103": {
+      "content": "[SEP]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "104": {
+      "content": "[MASK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "clean_up_tokenization_spaces": true,
+  "cls_token": "[CLS]",
+  "do_lower_case": true,
+  "mask_token": "[MASK]",
+  "max_length": 256,
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_to_multiple_of": null,
+  "pad_token": "[PAD]",
+  "pad_token_type_id": 0,
+  "padding_side": "right",
+  "sep_token": "[SEP]",
+  "stride": 0,
+  "strip_accents": true,
+  "tokenize_chinese_chars": true,
+  "tokenizer_class": "BertTokenizer",
+  "truncation_side": "right",
+  "truncation_strategy": "longest_first",
+  "unk_token": "[UNK]"
+}

checkpoint-1/trainer_state.json

@@ -0,0 +1,40 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.0,
+  "eval_steps": 10000,
+  "global_step": 1,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "grad_norm": 4.18489933013916,
+      "learning_rate": 0.0,
+      "loss": 0.8612,
+      "step": 1
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 1,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 15000,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 21821285850.0,
+  "train_batch_size": 32,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-1/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:253c832ed6cc1e1c6e4407d924248dfb9827691f8ff7f7c2cbbb03eb7f1936d7
+size 5841

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "/workspaces/prod/nlp/nlp-tools/data/ner_demo_de/models/tinybert-clinalytix",
+  "_name_or_path": "DedalusHealthCare/tinybert-mlm-de",
   "architectures": [
     "BertForTokenClassification"
   ],
@@ -10,14 +10,20 @@
   "hidden_dropout_prob": 0.1,
   "hidden_size": 312,
   "id2label": {
-    "0": "B-DISORDER_FINDING",
-    "1": "O"
+    "0": "B-DISORDER",
+    "1": "B-FINDING",
+    "2": "I-DISORDER",
+    "3": "I-FINDING",
+    "4": "O"
   },
   "initializer_range": 0.02,
   "intermediate_size": 312,
   "label2id": {
-    "B-DISORDER_FINDING": 0,
-    "O": 1
+    "B-DISORDER": 0,
+    "B-FINDING": 1,
+    "I-DISORDER": 2,
+    "I-FINDING": 3,
+    "O": 4
   },
   "layer_norm_eps": 1e-12,
   "max_position_embeddings": 512,
@@ -27,6 +33,7 @@
   "pad_token_id": 0,
   "position_embedding_type": "absolute",
   "pre_trained": "",
+  "torch_dtype": "float32",
   "training": "",
   "transformers_version": "4.45.1",
   "type_vocab_size": 2,

eval_results.json

@@ -0,0 +1,20 @@
+{
+    "epoch": 1.0,
+    "eval_DISORDER_f1": 0.0,
+    "eval_DISORDER_number": 2,
+    "eval_DISORDER_precision": 0.0,
+    "eval_DISORDER_recall": 0.0,
+    "eval_FINDING_f1": 0.0,
+    "eval_FINDING_number": 0,
+    "eval_FINDING_precision": 0.0,
+    "eval_FINDING_recall": 0.0,
+    "eval_loss": 1.7064510583877563,
+    "eval_overall_accuracy": 0.11764705882352941,
+    "eval_overall_f1": 0.0,
+    "eval_overall_precision": 0.0,
+    "eval_overall_recall": 0.0,
+    "eval_runtime": 0.1739,
+    "eval_samples": 3,
+    "eval_samples_per_second": 17.256,
+    "eval_steps_per_second": 5.752
+}

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8a637edcc6d6ebfd24fdd2df56ffec67e4170ffc3a5a07a186b712b100a03f9f
-size 48864824
+oid sha256:70bb4470bd4149e9aa0d4ab934e0e4ffc533b8af1702686320f4b3d57f8a062e
+size 48868580

model_info.json

@@ -1,16 +1,16 @@
 {
-    "model_version": 1768930002,
-    "model_name": "bert-demo-de",
-    "model_type": "bert",
+    "model_version": 1774364768,
+    "model_name": "tinybert-demo-de",
+    "model_type": "tinybert",
     "model_platform": "pytorch",
-    "model_architecture": "BERT",
+    "model_architecture": "TinyBERT",
     "model_description": "Retrieve named entities from text.",
-    "model_date": "2026-01-20T18:26:42.445432+01:00",
-    "clinalytix_version": "unknown",
+    "model_date": "2026-03-24T15:06:08.594093+00:00",
+    "clinalytix_version": "26.03.0",
     "model_objective": "RECOGNITION",
     "use_case": "demo",
-    "build_number": null,
-    "revision_number": null,
+    "build_number": "10",
+    "revision_number": "7a69bd200ca16eb3f14e380484a5fb61afc70893",
     "language_code": "de",
     "language_codes_multilingual": null,
     "target": null,

runs/Mar24_12-30-41_ip-10-246-1-57/events.out.tfevents.1774355449.ip-10-246-1-57.121762.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9da6d43d0ed82cd545e671a6d07a7b8dded3014c3cf749d1944564cfab90944f
+size 6072

runs/Mar24_12-30-41_ip-10-246-1-57/events.out.tfevents.1774355450.ip-10-246-1-57.121762.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:13b01f02d5d3b877434f77fe121a102a0f2e3a814abbed9e98729abb747f0b8b
+size 1041

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 1.0,
+    "total_flos": 21821285850.0,
+    "train_loss": 0.8611775636672974,
+    "train_runtime": 1.1366,
+    "train_samples": 17,
+    "train_samples_per_second": 14.957,
+    "train_steps_per_second": 0.88
+}

trainer_state.json

@@ -0,0 +1,49 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.0,
+  "eval_steps": 10000,
+  "global_step": 1,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "grad_norm": 4.18489933013916,
+      "learning_rate": 0.0,
+      "loss": 0.8612,
+      "step": 1
+    },
+    {
+      "epoch": 1.0,
+      "step": 1,
+      "total_flos": 21821285850.0,
+      "train_loss": 0.8611775636672974,
+      "train_runtime": 1.1366,
+      "train_samples_per_second": 14.957,
+      "train_steps_per_second": 0.88
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 1,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 15000,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 21821285850.0,
+  "train_batch_size": 32,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:253c832ed6cc1e1c6e4407d924248dfb9827691f8ff7f7c2cbbb03eb7f1936d7
+size 5841