Update app.py

app.py

@@ -1,64 +1,91 @@
+import os
+import json
+import torch
+import pandas as pd
+import numpy as np
+from transformers import AutoTokenizer, AutoModel
+from sklearn.linear_model import LogisticRegression
+from sklearn.preprocessing import LabelEncoder
 import gradio as gr
-from huggingface_hub import InferenceClient
 
-"""
-For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
-"""
-client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")
+# Load the model and tokenizer (Bio_ClinicalBERT)
+model_name = "emilyalsentzer/Bio_ClinicalBERT"
+tokenizer = AutoTokenizer.from_pretrained(model_name)
+model = AutoModel.from_pretrained(model_name)
+model.eval()
 
+# Load disease mapping
+with open("disease_mapping.json", "r") as f:
+    disease_data = json.load(f)
+    disease_info = {item["Disease"].lower(): item for item in disease_data}
 
-def respond(
-    message,
-    history: list[tuple[str, str]],
-    system_message,
-    max_tokens,
-    temperature,
-    top_p,
-):
-    messages = [{"role": "system", "content": system_message}]
+# Load precomputed embeddings
+train_embeddings = pd.read_csv("train_embeddings.csv")
+test_embeddings = pd.read_csv("test_embeddings.csv")
 
-    for val in history:
-        if val[0]:
-            messages.append({"role": "user", "content": val[0]})
-        if val[1]:
-            messages.append({"role": "assistant", "content": val[1]})
+# Encode disease labels
+le = LabelEncoder()
+train_embeddings["label"] = le.fit_transform(train_embeddings["label"])
+test_embeddings["label"] = le.transform(test_embeddings["label"])
 
-    messages.append({"role": "user", "content": message})
+# Split X and y for training
+X_train = train_embeddings.drop(columns=["label"]).values
+y_train = train_embeddings["label"].values
+X_test = test_embeddings.drop(columns=["label"]).values
+y_test = test_embeddings["label"].values
 
-    response = ""
+# Train Logistic Regression classifier
+clf = LogisticRegression(max_iter=1000)
+clf.fit(X_train, y_train)
 
-    for message in client.chat_completion(
-        messages,
-        max_tokens=max_tokens,
-        stream=True,
-        temperature=temperature,
-        top_p=top_p,
-    ):
-        token = message.choices[0].delta.content
+# Prediction function
+def predict_disease(symptoms):
+    """
+    Predicts the top 3 diseases based on the input symptoms using the trained Logistic Regression model
+    and provides disease information.
 
-        response += token
-        yield response
+    Args:
+        symptoms (str): The input symptoms.
+    
+    Returns:
+        list: A list of dictionaries, each containing information about a predicted disease.
+    """
+    emb = get_embedding(symptoms).reshape(1, -1)
+    probs = clf.predict_proba(emb)[0]
+    top3_idx = np.argsort(probs)[::-1][:3]
+    results = []
+    for idx in top3_idx:
+        disease = le.inverse_transform([idx])[0]
+        info = disease_info.get(disease.lower(), {})
+        results.append({
+            "Disease": disease,
+            "Confidence": round(probs[idx]*100, 2),
+            "Description": info.get("Description", "Not available"),
+            "Severity": info.get("Severity", "Not available"),
+            "Precaution": info.get("Precaution", "Not available")
+        })
+    return results
 
+# Gradio UI function
+def chatbot(symptoms):
+    """
+    Gradio interface function for getting disease predictions based on symptoms.
+    """
+    preds = predict_disease(symptoms)
+    output = ""
+    for i, pred in enumerate(preds, 1):
+        output += f"### Prediction {i}\n"
+        output += f"- **Disease**: {pred['Disease']} ({pred['Confidence']}%)\n"
+        output += f"- **Description**: {pred['Description']}\n"
+        output += f"- **Severity**: {pred['Severity']}\n"
+        output += f"- **Precaution**: {pred['Precaution']}\n\n"
+    return output.strip()
 
-"""
-For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
-"""
-demo = gr.ChatInterface(
-    respond,
-    additional_inputs=[
-        gr.Textbox(value="You are a friendly Chatbot.", label="System message"),
-        gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
-        gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
-        gr.Slider(
-            minimum=0.1,
-            maximum=1.0,
-            value=0.95,
-            step=0.05,
-            label="Top-p (nucleus sampling)",
-        ),
-    ],
-)
-
-
-if __name__ == "__main__":
-    demo.launch()
+# Start the Gradio interface
+gr.Interface(
+    fn=chatbot,
+    inputs=gr.Textbox(label="Enter your symptoms"),
+    outputs=gr.Markdown(),
+    title="🩺 BioClinicalBERT Medical Chatbot",
+    description="Enter your symptoms to receive top 3 disease predictions with severity and precautions."
+).launch()