Create app.py

app.py

@@ -0,0 +1,269 @@
+import requests
+from transformers import pipeline
+import requests
+from transformers import pipeline
+import requests
+from transformers import pipeline
+from pymongo import MongoClient
+import datetime
+from flask import Flask, request, jsonify
+
+# Initialize Flask app
+app = Flask(__name__)
+
+# Set up Hugging Face model for conversational AI
+nlp = pipeline("conversational", model="microsoft/DialoGPT-medium")
+
+# Set up MongoDB client for storing conversation history
+client = MongoClient("mongodb://localhost:27017/")  # Use your MongoDB URI if cloud-based
+db = client.healthbot
+
+# Function to fetch medical information using an API
+def get_medical_info(query):
+    # Example API call (replace with real API or local dataset)
+    api_url = f'https://api.healthapi.com/v1/query/{query}'
+    headers = {'Authorization': 'Bearer YOUR_API_KEY'}  # Replace with your API key
+    
+    try:
+        response = requests.get(api_url, headers=headers)
+        if response.status_code == 200:
+            return response.json()  # Example: {'description': 'Details about the symptom'}
+        else:
+            return {'error': 'Could not retrieve information from API.'}
+    except Exception as e:
+        return {'error': str(e)}
+
+# Save conversations to MongoDB
+def save_conversation(user_input, bot_response):
+    conversation = {
+        'timestamp': datetime.datetime.now(),
+        'user_input': user_input,
+        'bot_response': bot_response
+    }
+    db.conversations.insert_one(conversation)
+
+# Get the last 5 conversations from memory
+def get_previous_conversations():
+    return db.conversations.find().sort("timestamp", -1).limit(5)
+
+# Main function to handle user input
+def handle_user_input(user_input):
+    # If the input contains the word 'symptom', fetch medical info
+    if 'symptom' in user_input.lower():
+        query = user_input.lower().replace('symptom', '')
+        medical_info = get_medical_info(query)
+        if 'error' in medical_info:
+            return f"Sorry, we couldn't find information about '{query}'. Please try again."
+        else:
+            return f"Based on your query '{query}', here is some information: {medical_info.get('description', 'No details found.')}"
+    else:
+        # Otherwise, handle general conversation
+        response = nlp(user_input)
+        return response[0]['generated_text']
+
+# Flask route to handle chatbot interaction
+@app.route('/chat', methods=['POST'])
+def chat():
+    user_input = request.json.get('user_input')
+    if not user_input:
+        return jsonify({'error': 'User input is required'}), 400
+    
+    # Get response from chatbot
+    bot_response = handle_user_input(user_input)
+    
+    # Save the conversation to memory
+    save_conversation(user_input, bot_response)
+    
+    return jsonify({'response': bot_response})
+
+# Flask route to retrieve previous conversations
+@app.route('/history', methods=['GET'])
+def history():
+    conversations = get_previous_conversations()
+    history = []
+    for convo in conversations:
+        history.append({
+            'timestamp': convo['timestamp'].strftime('%Y-%m-%d %H:%M:%S'),
+            'user_input': convo['user_input'],
+            'bot_response': convo['bot_response']
+        })
+    return jsonify(history)
+
+# Run the Flask app
+if __name__ == "__main__":
+    app.run(debug=True)
+
+import datetime
+from flask import Flask, request, jsonify
+
+# Initialize Flask app
+app = Flask(__name__)
+
+# Set up Hugging Face model for conversational AI
+nlp = pipeline("conversational", model="microsoft/DialoGPT-medium")
+
+# Set up MongoDB client for storing conversation history
+client = MongoClient("mongodb://localhost:27017/")  # Use your MongoDB URI if cloud-based
+db = client.healthbot
+
+# Function to fetch medical information using an API
+def get_medical_info(query):
+    # Example API call (replace with real API or local dataset)
+    api_url = f'https://api.healthapi.com/v1/query/{query}'
+    headers = {'Authorization': 'Bearer YOUR_API_KEY'}  # Replace with your API key
+    
+    try:
+        response = requests.get(api_url, headers=headers)
+        if response.status_code == 200:
+            return response.json()  # Example: {'description': 'Details about the symptom'}
+        else:
+            return {'error': 'Could not retrieve information from API.'}
+    except Exception as e:
+        return {'error': str(e)}
+
+# Save conversations to MongoDB
+def save_conversation(user_input, bot_response):
+    conversation = {
+        'timestamp': datetime.datetime.now(),
+        'user_input': user_input,
+        'bot_response': bot_response
+    }
+    db.conversations.insert_one(conversation)
+
+# Get the last 5 conversations from memory
+def get_previous_conversations():
+    return db.conversations.find().sort("timestamp", -1).limit(5)
+
+# Main function to handle user input
+def handle_user_input(user_input):
+    # If the input contains the word 'symptom', fetch medical info
+    if 'symptom' in user_input.lower():
+        query = user_input.lower().replace('symptom', '')
+        medical_info = get_medical_info(query)
+        if 'error' in medical_info:
+            return f"Sorry, we couldn't find information about '{query}'. Please try again."
+        else:
+            return f"Based on your query '{query}', here is some information: {medical_info.get('description', 'No details found.')}"
+    else:
+        # Otherwise, handle general conversation
+        response = nlp(user_input)
+        return response[0]['generated_text']
+
+# Flask route to handle chatbot interaction
+@app.route('/chat', methods=['POST'])
+def chat():
+    user_input = request.json.get('user_input')
+    if not user_input:
+        return jsonify({'error': 'User input is required'}), 400
+    
+    # Get response from chatbot
+    bot_response = handle_user_input(user_input)
+    
+    # Save the conversation to memory
+    save_conversation(user_input, bot_response)
+    
+    return jsonify({'response': bot_response})
+
+# Flask route to retrieve previous conversations
+@app.route('/history', methods=['GET'])
+def history():
+    conversations = get_previous_conversations()
+    history = []
+    for convo in conversations:
+        history.append({
+            'timestamp': convo['timestamp'].strftime('%Y-%m-%d %H:%M:%S'),
+            'user_input': convo['user_input'],
+            'bot_response': convo['bot_response']
+        })
+    return jsonify(history)
+
+# Run the Flask app
+if __name__ == "__main__":
+    app.run(debug=True)
+import datetime
+from flask import Flask, request, jsonify
+
+# Initialize Flask app
+app = Flask(__name__)
+
+# Set up Hugging Face model for conversational AI
+nlp = pipeline("conversational", model="microsoft/DialoGPT-medium")
+
+# Set up MongoDB client for storing conversation history
+client = MongoClient("mongodb://localhost:27017/")  # Use your MongoDB URI if cloud-based
+db = client.healthbot
+
+# Function to fetch medical information using an API
+def get_medical_info(query):
+    # Example API call (replace with real API or local dataset)
+    api_url = f'https://api.healthapi.com/v1/query/{query}'
+    headers = {'Authorization': 'Bearer YOUR_API_KEY'}  # Replace with your API key
+    
+    try:
+        response = requests.get(api_url, headers=headers)
+        if response.status_code == 200:
+            return response.json()  # Example: {'description': 'Details about the symptom'}
+        else:
+            return {'error': 'Could not retrieve information from API.'}
+    except Exception as e:
+        return {'error': str(e)}
+
+# Save conversations to MongoDB
+def save_conversation(user_input, bot_response):
+    conversation = {
+        'timestamp': datetime.datetime.now(),
+        'user_input': user_input,
+        'bot_response': bot_response
+    }
+    db.conversations.insert_one(conversation)
+
+# Get the last 5 conversations from memory
+def get_previous_conversations():
+    return db.conversations.find().sort("timestamp", -1).limit(5)
+
+# Main function to handle user input
+def handle_user_input(user_input):
+    # If the input contains the word 'symptom', fetch medical info
+    if 'symptom' in user_input.lower():
+        query = user_input.lower().replace('symptom', '')
+        medical_info = get_medical_info(query)
+        if 'error' in medical_info:
+            return f"Sorry, we couldn't find information about '{query}'. Please try again."
+        else:
+            return f"Based on your query '{query}', here is some information: {medical_info.get('description', 'No details found.')}"
+    else:
+        # Otherwise, handle general conversation
+        response = nlp(user_input)
+        return response[0]['generated_text']
+
+# Flask route to handle chatbot interaction
+@app.route('/chat', methods=['POST'])
+def chat():
+    user_input = request.json.get('user_input')
+    if not user_input:
+        return jsonify({'error': 'User input is required'}), 400
+    
+    # Get response from chatbot
+    bot_response = handle_user_input(user_input)
+    
+    # Save the conversation to memory
+    save_conversation(user_input, bot_response)
+    
+    return jsonify({'response': bot_response})
+
+# Flask route to retrieve previous conversations
+@app.route('/history', methods=['GET'])
+def history():
+    conversations = get_previous_conversations()
+    history = []
+    for convo in conversations:
+        history.append({
+            'timestamp': convo['timestamp'].strftime('%Y-%m-%d %H:%M:%S'),
+            'user_input': convo['user_input'],
+            'bot_response': convo['bot_response']
+        })
+    return jsonify(history)
+
+# Run the Flask app
+if __name__ == "__main__":
+    app.run(debug=True)