Spaces:
Configuration error
Configuration error
File size: 4,834 Bytes
bec06d9 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 |
from flask import Flask, request, jsonify
from flask_cors import CORS
import sys
import os
# Add the src/python directory to the path so we can import our modules
sys.path.insert(0, os.path.join(os.path.dirname(__file__)))
sys.path.insert(0, os.path.join(os.path.dirname(__file__), 'src', 'python'))
from src.python.vector_store import VectorStore
from embedder_wrapper import SyncEmbedder
from src.python.config import OPENAI_API_KEY, QDRANT_URL, QDRANT_API_KEY, COLLECTION_NAME
import logging
# Configure logging
logging.basicConfig(level=logging.INFO)
logger = logging.getLogger(__name__)
app = Flask(__name__)
CORS(app, origins=["http://localhost:3000", "http://localhost:5000", "http://127.0.0.1:3000", "http://127.0.0.1:5000"], supports_credentials=True, allow_headers=["Content-Type", "Authorization"])
# Initialize our components
try:
vector_store = VectorStore()
embedder = SyncEmbedder()
logger.info("Service initialized successfully")
logger.info(f"Connected to Qdrant at: {QDRANT_URL}")
except Exception as e:
logger.error(f"Failed to initialize services: {str(e)}")
raise
@app.route('/chat', methods=['POST'])
def chat():
"""
Chat endpoint that takes user query and returns a response based on RAG
Expected JSON format: {"message": "user question"}
"""
try:
data = request.get_json()
if not data or 'message' not in data:
return jsonify({'error': 'Message field is required'}), 400
user_message = data['message']
# Create embedding for the user message
try:
query_embedding = embedder.embed_text(user_message)
except Exception as e:
logger.error(f"Error creating embedding: {str(e)}")
return jsonify({'error': f'Error processing your message: {str(e)}'}), 500
# Search for similar documents in the vector store
try:
similar_docs = vector_store.search_similar(query_embedding, top_k=5)
except Exception as e:
logger.error(f"Error searching for documents: {str(e)}")
return jsonify({'error': f'Error retrieving documents: {str(e)}'}), 500
# Format the retrieved documents as context
context = "\n".join([doc['content'] for doc in similar_docs])
# Prepare the prompt for the LLM
if context.strip() == "":
# If no context is found, let the model respond without it
prompt = f"""
Please answer the following question. If you don't know the answer, please say so.
Question: {user_message}
Answer:
"""
else:
prompt = f"""
Answer the question based on the context provided.
If the answer is not in the context, say "I don't have enough information to answer that question."
Context: {context}
Question: {user_message}
Answer:
"""
# Use OpenAI API to generate the response
try:
from openai import OpenAI
client = OpenAI(api_key=OPENAI_API_KEY)
response = client.chat.completions.create(
model="gpt-3.5-turbo",
messages=[{"role": "user", "content": prompt}],
temperature=0.3,
max_tokens=500
)
bot_response = response.choices[0].message.content.strip()
except Exception as e:
logger.error(f"Error calling LLM: {str(e)}")
return jsonify({'error': f'Error generating response: {str(e)}'}), 500
return jsonify({
'response': bot_response,
'sources': [doc.get('source', '') for doc in similar_docs],
'scores': [doc.get('score', 0.0) for doc in similar_docs],
'retrieved_context': context
})
except Exception as e:
logger.error(f"Unexpected error in chat endpoint: {str(e)}")
return jsonify({'error': str(e)}), 500
@app.route('/health', methods=['GET'])
def health():
"""Health check endpoint"""
return jsonify({'status': 'healthy'})
@app.route('/documents/count', methods=['GET'])
def document_count():
"""Get the count of documents in the vector store"""
try:
count = vector_store.get_all_documents_count()
return jsonify({'count': count})
except Exception as e:
logger.error(f"Error getting document count: {str(e)}")
return jsonify({'error': str(e)}), 500
# Hugging Face Spaces compatible application entry point
if __name__ != '__main__':
# When run as part of Hugging Face Spaces
import os
port = int(os.environ.get('PORT', 5000))
app.run(debug=False, host='0.0.0.0', port=port)
if __name__ == '__main__':
# For local development
app.run(debug=True, host='0.0.0.0', port=5000) |