|
|
--- |
|
|
title: SVM ADR Severity Classifier 2 |
|
|
emoji: 🌍 |
|
|
colorFrom: blue |
|
|
colorTo: purple |
|
|
sdk: gradio |
|
|
sdk_version: 5.25.2 |
|
|
app_file: app.py |
|
|
pinned: false |
|
|
--- |
|
|
|
|
|
# ADR Severity Classifier |
|
|
|
|
|
This is a binary text classification demo using a fine-tuned `roberta-base` model. It predicts whether an adverse drug reaction (ADR) is **severe** or **not severe** based on a short textual description. |
|
|
|
|
|
### Model |
|
|
- Based on `roberta-base`, fine-tuned on biomedical ADR descriptions |
|
|
- Prioritises high recall on severe cases (class 1) |
|
|
|
|
|
### Try it |
|
|
Type or paste a drug reaction report (e.g., "Severe headache and shortness of breath after taking...") and the model will return probabilities for each class. |
|
|
|
|
|
Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference |
|
|
|
