Update README.md

README.md

@@ -1,60 +1,90 @@
 ---
-library_name: transformers
+language: en
 license: apache-2.0
 base_model: distilbert-base-uncased
+library_name: transformers
 tags:
-- generated_from_trainer
+- medical
+- ner
+- token-classification
+- healthcare
+- clinical-nlp
+datasets:
+- sohaibdevv/medical-prescription-ner-2026-benchmark
+metrics:
+- loss
+pipeline_tag: token-classification
 model-index:
 - name: Medical-NER-2026-Success
-  results: []
+  results:
+  - task:
+      type: token-classification
+      name: Named Entity Recognition
+    dataset:
+      name: Medical Prescription NER 2026 Benchmark
+      type: csv
+    metrics:
+    - type: loss
+      value: 0.5459
+      name: Validation Loss
+widget:
+- text: "Take 500mg of Amoxicillin twice daily for 7 days."
+  example_title: "Standard Prescription"
+- text: "Administer 10ml of Ibuprofen at night."
+  example_title: "Liquid Dosage"
 ---
 
-<!-- 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. -->
-
 # Medical-NER-2026-Success
 
-This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on the None dataset.
-It achieves the following results on the evaluation set:
-- Loss: 0.5459
-
-## Model description
-
-More information needed
-
-## Intended uses & limitations
+## Overview
+This model is a specialized **Named Entity Recognition (NER)** tool fine-tuned from **DistilBERT**. It is specifically designed to extract clinical entities from medical prescriptions and doctor notes. This project was developed as a benchmark for 2026 Medical NLP tasks.
 
-More information needed
+### Detected Entities
+| Label | Description | Example |
+| :--- | :--- | :--- |
+| **DRUG** | Name of the medication | *Aspirin, Insulin, Amoxicillin* |
+| **DOSAGE** | Amount, strength, or form | *500mg, 2 tablets, 10ml* |
+| **FREQ** | Frequency and timing | *Daily, twice a day, every 8 hours* |
 
-## Training and evaluation data
+## How to use
+You can use this model directly with the Hugging Face `pipeline`:
 
-More information needed
+```python
+from transformers import pipeline
 
-## Training procedure
+# Load the model
+ner_pipe = pipeline("token-classification", 
+                    model="sohaibdevv/Medical-NER-2026-Success", 
+                    aggregation_strategy="simple")
 
-### Training hyperparameters
+# Test a prescription
+text = "Patient is prescribed 20mg of Lisinopril once daily."
+results = ner_pipe(text)
 
-The following hyperparameters were used during training:
-- learning_rate: 2e-05
-- train_batch_size: 8
-- eval_batch_size: 8
-- seed: 42
-- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
-- lr_scheduler_type: linear
-- num_epochs: 3
+for entity in results:
+    print(f"Entity: {entity['word']} | Label: {entity['entity_group']}")
+```
 
-### Training results
+## Training Details
+The model was trained using a **Rule-Based Bootstrapping** approach on the 2026 Medical Benchmark dataset. 
 
-| Training Loss | Epoch | Step | Validation Loss |
-|:-------------:|:-----:|:----:|:---------------:|
-| No log        | 1.0   | 4    | 1.1384          |
-| No log        | 2.0   | 8    | 0.7012          |
-| 1.1499        | 3.0   | 12   | 0.5459          |
+* **Base Model:** `distilbert-base-uncased`
+* **Labels:** 7 (BIO format for Drug, Dosage, and Frequency)
+* **Epochs:** 3
+* **Learning Rate:** 2e-05
+* **Optimization:** AdamW with linear scheduler
 
+### Performance
+The model achieved a **Validation Loss of 0.5459**, showing strong convergence for medical entity detection in structured English sentences.
 
-### Framework versions
+## Limitations & Ethics
+- **Research Only:** This model is for educational and research purposes. 
+- **Not for Diagnosis:** It should never be used to automate clinical decisions without professional human oversight.
+- **English Only:** Currently optimized for English-language medical text.
 
+## Framework Versions
 - Transformers 5.0.0
 - Pytorch 2.10.0+cpu
 - Datasets 4.8.3
 - Tokenizers 0.22.2
+```