feat: Upload fine-tuned medical NER model OpenMed-NER-DiseaseDetect-SuperClinical-184M

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* 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
+openmed_vs_sota_grouped_bars.png filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,242 @@
+---
+widget:
+- text: "The patient was diagnosed with diabetes mellitus type 2."
+- text: "Symptoms of Alzheimer's disease became apparent over several months."
+- text: "Treatment for hypertension was initiated immediately."
+- text: "A possible link between Crohn's disease and gut microbiota is being investigated."
+- text: "The patient has a family history of cystic fibrosis."
+tags:
+- token-classification
+- named-entity-recognition
+- biomedical-nlp
+- transformers
+- disease-entity-recognition
+- medical-diagnosis
+- pathology
+- biocuration
+- disease
+language:
+- en
+license: apache-2.0
+---
+
+# 🧬 [OpenMed-NER-DiseaseDetect-SuperClinical-184M](https://huggingface.co/OpenMed/OpenMed-NER-DiseaseDetect-SuperClinical-184M)
+
+**Specialized model for Disease Entity Recognition - Disease entities from the BC5CDR dataset**
+
+[![License](https://img.shields.io/badge/License-Apache%202.0-blue.svg)](https://opensource.org/licenses/Apache-2.0)
+[![Python](https://img.shields.io/badge/Python-3.8%2B-blue)]()
+[![Transformers](https://img.shields.io/badge/🤗-Transformers-yellow)]()
+[![OpenMed](https://img.shields.io/badge/🏥-OpenMed-green)](https://huggingface.co/OpenMed)
+
+## 📋 Model Overview
+
+This model is a **state-of-the-art** fine-tuned transformer engineered to deliver **enterprise-grade accuracy** for disease entity recognition - disease entities from the bc5cdr dataset. This specialized model excels at identifying and extracting biomedical entities from clinical texts, research papers, and healthcare documents, enabling applications such as **drug interaction detection**, **medication extraction from patient records**, **adverse event monitoring**, **literature mining for drug discovery**, and **biomedical knowledge graph construction** with **production-ready reliability** for clinical and research applications.
+
+### 🎯 Key Features
+- **High Precision**: Optimized for biomedical entity recognition
+- **Domain-Specific**: Trained on curated BC5CDR_DISEASE dataset
+- **Production-Ready**: Validated on clinical benchmarks
+- **Easy Integration**: Compatible with Hugging Face Transformers ecosystem
+
+### 🏷️ Supported Entity Types
+
+This model can identify and classify the following biomedical entities:
+
+- `B-DISEASE`
+- `I-DISEASE`
+
+## 📊 Dataset
+
+BC5CDR-Disease targets disease entity recognition from the BioCreative V Chemical-Disease Relation extraction corpus.
+
+The BC5CDR-Disease corpus is the disease-focused component of the BioCreative V Chemical-Disease Relation (CDR) task, containing 1,500 PubMed abstracts with 5,818 annotated disease entities. This manually curated dataset is designed to advance automated disease name recognition for medical diagnosis, pathology research, and clinical decision support systems. The corpus includes annotations for various disease types, medical conditions, and pathological states mentioned in biomedical literature. It serves as a benchmark for evaluating NER models in clinical and biomedical applications where accurate disease entity identification is crucial for medical informatics and healthcare analytics.
+
+
+## 📊 Performance Metrics
+
+### Current Model Performance
+- **F1 Score**: `0.89`
+- **Precision**: `0.87`
+- **Recall**: `0.92`
+- **Accuracy**: `0.98`
+
+### 🏆 Comparative Performance on BC5CDR_DISEASE Dataset
+
+| Rank | Model | F1 Score | Precision | Recall | Accuracy |
+|------|-------|----------|-----------|--------|-----------|
+| 🥇 1 | [OpenMed-NER-DiseaseDetect-SuperClinical-434M](https://huggingface.co/OpenMed/OpenMed-NER-DiseaseDetect-SuperClinical-434M) | **0.9118** | 0.9028 | 0.9211 | 0.9839 |
+| 🥈 2 | [OpenMed-NER-DiseaseDetect-PubMed-335M](https://huggingface.co/OpenMed/OpenMed-NER-DiseaseDetect-PubMed-335M) | **0.9097** | 0.8932 | 0.9268 | 0.9849 |
+| 🥉 3 | [OpenMed-NER-DiseaseDetect-MultiMed-335M](https://huggingface.co/OpenMed/OpenMed-NER-DiseaseDetect-MultiMed-335M) | **0.9022** | 0.8890 | 0.9159 | 0.9758 |
+|  4 | [OpenMed-NER-DiseaseDetect-BioMed-335M](https://huggingface.co/OpenMed/OpenMed-NER-DiseaseDetect-BioMed-335M) | **0.9005** | 0.8887 | 0.9126 | 0.9838 |
+|  5 | [OpenMed-NER-DiseaseDetect-BioClinical-108M](https://huggingface.co/OpenMed/OpenMed-NER-DiseaseDetect-BioClinical-108M) | **0.8999** | 0.8862 | 0.9140 | 0.9723 |
+|  6 | [OpenMed-NER-DiseaseDetect-PubMed-109M](https://huggingface.co/OpenMed/OpenMed-NER-DiseaseDetect-PubMed-109M) | **0.8994** | 0.8899 | 0.9091 | 0.9839 |
+|  7 | [OpenMed-NER-DiseaseDetect-BioPatient-108M](https://huggingface.co/OpenMed/OpenMed-NER-DiseaseDetect-BioPatient-108M) | **0.8991** | 0.8864 | 0.9121 | 0.9721 |
+|  8 | [OpenMed-NER-DiseaseDetect-SuperClinical-184M](https://huggingface.co/OpenMed/OpenMed-NER-DiseaseDetect-SuperClinical-184M) | **0.8943** | 0.8687 | 0.9214 | 0.9812 |
+|  9 | [OpenMed-NER-DiseaseDetect-SuperClinical-141M](https://huggingface.co/OpenMed/OpenMed-NER-DiseaseDetect-SuperClinical-141M) | **0.8921** | 0.8686 | 0.9170 | 0.9809 |
+|  10 | [OpenMed-NER-DiseaseDetect-MultiMed-568M](https://huggingface.co/OpenMed/OpenMed-NER-DiseaseDetect-MultiMed-568M) | **0.8909** | 0.8803 | 0.9017 | 0.9776 |
+
+
+*Rankings based on F1-score performance across all models trained on this dataset.*
+
+![OpenMed (open-source) vs. latest closed-source SOTA](https://huggingface.co/spaces/OpenMed/README/resolve/main/openmed_vs_sota_performance.png)
+
+*Figure: OpenMed (Open-Source) vs. Latest SOTA (Closed-Source) performance comparison across biomedical NER datasets.*
+
+## 🚀 Quick Start
+
+### Installation
+
+```bash
+pip install transformers torch
+```
+
+### Usage
+
+```python
+from transformers import pipeline
+
+# Load the model and tokenizer
+# Model: https://huggingface.co/OpenMed/OpenMed-NER-DiseaseDetect-SuperClinical-184M
+model_name = "OpenMed/OpenMed-NER-DiseaseDetect-SuperClinical-184M"
+
+# Create a pipeline
+medical_ner_pipeline = pipeline(
+    model=model_name,
+    aggregation_strategy="simple"
+)
+
+# Example usage
+text = "The patient was diagnosed with diabetes mellitus type 2."
+entities = medical_ner_pipeline(text)
+
+print(entities)
+
+token = entities[0]
+print(text[token["start"] : token["end"]])
+```
+
+NOTE: The `aggregation_strategy` parameter defines how token predictions are grouped into entities. For a detailed explanation, please refer to the [Hugging Face documentation](https://huggingface.co/docs/transformers/en/main_classes/pipelines#transformers.TokenClassificationPipeline.aggregation_strategy).
+
+Here is a summary of the available strategies:
+- **`none`**: Returns raw token predictions without any aggregation.
+- **`simple`**: Groups adjacent tokens with the same entity type (e.g., `B-LOC` followed by `I-LOC`).
+- **`first`**: For word-based models, if tokens within a word have different entity tags, the tag of the first token is assigned to the entire word.
+- **`average`**: For word-based models, this strategy averages the scores of tokens within a word and applies the label with the highest resulting score.
+- **`max`**: For word-based models, the entity label from the token with the highest score within a word is assigned to the entire word.
+
+### Batch Processing
+
+For efficient processing of large datasets, use proper batching with the `batch_size` parameter:
+
+```python
+texts = [
+    "The patient was diagnosed with diabetes mellitus type 2.",
+    "Symptoms of Alzheimer's disease became apparent over several months.",
+    "Treatment for hypertension was initiated immediately.",
+    "A possible link between Crohn's disease and gut microbiota is being investigated.",
+    "The patient has a family history of cystic fibrosis.",
+]
+
+# Efficient batch processing with optimized batch size
+# Adjust batch_size based on your GPU memory (typically 8, 16, 32, or 64)
+results = medical_ner_pipeline(texts, batch_size=8)
+
+for i, entities in enumerate(results):
+    print(f"Text {i+1} entities:")
+    for entity in entities:
+        print(f"  - {entity['word']} ({entity['entity_group']}): {entity['score']:.4f}")
+```
+
+### Large Dataset Processing
+
+For processing large datasets efficiently:
+
+```python
+from transformers.pipelines.pt_utils import KeyDataset
+from datasets import Dataset
+import pandas as pd
+
+# Load your data
+# Load a medical dataset from Hugging Face
+from datasets import load_dataset
+
+# Load a public medical dataset (using a subset for testing)
+medical_dataset = load_dataset("BI55/MedText", split="train[:100]")  # Load first 100 examples
+data = pd.DataFrame({"text": medical_dataset["Completion"]})
+dataset = Dataset.from_pandas(data)
+
+# Process with optimal batching for your hardware
+batch_size = 16  # Tune this based on your GPU memory
+results = []
+
+for out in medical_ner_pipeline(KeyDataset(dataset, "text"), batch_size=batch_size):
+    results.extend(out)
+
+print(f"Processed {len(results)} texts with batching")
+
+```
+
+### Performance Optimization
+
+**Batch Size Guidelines:**
+- **CPU**: Start with batch_size=1-4
+- **Single GPU**: Try batch_size=8-32 depending on GPU memory
+- **High-end GPU**: Can handle batch_size=64 or higher
+- **Monitor GPU utilization** to find the optimal batch size for your hardware
+
+**Memory Considerations:**
+```python
+# For limited GPU memory, use smaller batches
+medical_ner_pipeline = pipeline(
+    model=model_name,
+    aggregation_strategy="simple",
+    device=0  # Specify GPU device
+)
+
+# Process with memory-efficient batching
+for batch_start in range(0, len(texts), batch_size):
+    batch = texts[batch_start:batch_start + batch_size]
+    batch_results = medical_ner_pipeline(batch, batch_size=len(batch))
+    results.extend(batch_results)
+```
+
+## 📚 Dataset Information
+
+- **Dataset**: BC5CDR_DISEASE
+- **Description**: Disease Entity Recognition - Disease entities from the BC5CDR dataset
+
+### Training Details
+- **Base Model**: deberta-v3-base
+- **Training Framework**: Hugging Face Transformers
+- **Optimization**: AdamW optimizer with learning rate scheduling
+- **Validation**: Cross-validation on held-out test set
+
+## 🔬 Model Architecture
+
+- **Base Architecture**: deberta-v3-base
+- **Task**: Token Classification (Named Entity Recognition)
+- **Labels**: Dataset-specific entity types
+- **Input**: Tokenized biomedical text
+- **Output**: BIO-tagged entity predictions
+
+## 💡 Use Cases
+
+This model is particularly useful for:
+- **Clinical Text Mining**: Extracting entities from medical records
+- **Biomedical Research**: Processing scientific literature
+- **Drug Discovery**: Identifying chemical compounds and drugs
+- **Healthcare Analytics**: Analyzing patient data and outcomes
+- **Academic Research**: Supporting biomedical NLP research
+
+## 📜 License
+
+Licensed under the Apache License 2.0. See [LICENSE](https://www.apache.org/licenses/LICENSE-2.0) for details.
+
+## 🤝 Contributing
+
+We welcome contributions of all kinds! Whether you have ideas, feature requests, or want to join our mission to advance open-source Healthcare AI, we'd love to hear from you.
+
+Follow [OpenMed Org](https://huggingface.co/OpenMed) on Hugging Face 🤗 and click "Watch" to stay updated on our latest releases and developments.
+
+

added_tokens.json

@@ -0,0 +1,3 @@
+{
+  "[MASK]": 128000
+}

config.json

@@ -0,0 +1,45 @@
+{
+  "architectures": [
+    "DebertaV2ForTokenClassification"
+  ],
+  "attention_probs_dropout_prob": 0.2,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.2,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "O",
+    "1": "B-DISEASE",
+    "2": "I-DISEASE"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "B-DISEASE": 1,
+    "I-DISEASE": 2,
+    "O": 0
+  },
+  "layer_norm_eps": 1e-07,
+  "legacy": true,
+  "max_position_embeddings": 512,
+  "max_relative_positions": -1,
+  "model_type": "deberta-v2",
+  "norm_rel_ebd": "layer_norm",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 0,
+  "pooler_dropout": 0,
+  "pooler_hidden_act": "gelu",
+  "pooler_hidden_size": 768,
+  "pos_att_type": [
+    "p2c",
+    "c2p"
+  ],
+  "position_biased_input": false,
+  "position_buckets": 256,
+  "relative_attention": true,
+  "share_att_key": true,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.53.2",
+  "type_vocab_size": 0,
+  "vocab_size": 128100
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a35dfc4030f67b6e6745938a75e0b6853ba5adec64494be6c14d9e748b300eba
+size 367692286

special_tokens_map.json

@@ -0,0 +1,15 @@
+{
+  "bos_token": "[CLS]",
+  "cls_token": "[CLS]",
+  "eos_token": "[SEP]",
+  "mask_token": "[MASK]",
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "unk_token": {
+    "content": "[UNK]",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  }
+}

spm.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c679fbf93643d19aab7ee10c0b99e460bdbc02fedf34b92b05af343b4af586fd
+size 2464616

test_results.json

@@ -0,0 +1,7 @@
+{
+    "eval_accuracy": 0.9811835106382979,
+    "eval_f1": 0.8942794528829442,
+    "eval_loss": 0.3396190106868744,
+    "eval_precision": 0.8687417413008369,
+    "eval_recall": 0.9213640610401744
+}

tokenizer_config.json

@@ -0,0 +1,59 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "[PAD]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "[CLS]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "[SEP]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "[UNK]",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128000": {
+      "content": "[MASK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "[CLS]",
+  "clean_up_tokenization_spaces": false,
+  "cls_token": "[CLS]",
+  "do_lower_case": false,
+  "eos_token": "[SEP]",
+  "extra_special_tokens": {},
+  "mask_token": "[MASK]",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "sp_model_kwargs": {},
+  "split_by_punct": false,
+  "tokenizer_class": "DebertaV2Tokenizer",
+  "unk_token": "[UNK]",
+  "vocab_type": "spm"
+}