update

app.py

@@ -19,23 +19,22 @@ app = FastAPI(
     version="1.0.0"
 )
 
-allowed_origins = os.getenv("ALLOWED_ORIGINS", "http://localhost:8000").split(",")
-
-# Add CORS middleware
+# Add CORS middleware - allow all origins for Hugging Face Spaces
 app.add_middleware(
     CORSMiddleware,
-    allow_origins=allowed_origins,  
+    allow_origins=["*"],  # Allow all origins in Hugging Face Spaces
     allow_credentials=True,
     allow_methods=["*"],
     allow_headers=["*"],
 )
 
-# Initialize chatbot
+# Initialize chatbot with Hugging Face Spaces specific settings
 chatbot = MentalHealthChatbot(
     model_name=os.getenv("MODEL_NAME", "meta-llama/Llama-3.2-3B-Instruct"),
-    peft_model_path=os.getenv("PEFT_MODEL_PATH", "llama_fine_tuned"),
+    peft_model_path=os.getenv("PEFT_MODEL_PATH", "nada013/mental-health-chatbot"),
     therapy_guidelines_path=os.getenv("GUIDELINES_PATH", "guidelines.txt"),
-    use_4bit=True
+    use_4bit=True,
+    device="cpu"  # Force CPU for Hugging Face Spaces
 )
 
 # pydantic models
@@ -194,4 +193,5 @@ async def get_user_replies(user_id: str):
 
 if __name__ == "__main__":
     import uvicorn
-    uvicorn.run(app, host="0.0.0.0", port=8000) 
+    # Use port 7860 for Hugging Face Spaces
+    uvicorn.run(app, host="0.0.0.0", port=7860) 

hf_spaces.py

@@ -0,0 +1,34 @@
+"""
+Hugging Face Spaces specific configuration
+"""
+import os
+
+# Set environment variables for Hugging Face Spaces
+os.environ.update({
+    'TRANSFORMERS_CACHE': '/tmp/huggingface',
+    'HF_HOME': '/tmp/huggingface',
+    'TOKENIZERS_PARALLELISM': 'false',
+    'TRANSFORMERS_VERBOSITY': 'error',
+    'BITSANDBYTES_NOWELCOME': '1'
+})
+
+# Create necessary directories
+for directory in ['/tmp/huggingface', '/tmp/vector_db', '/tmp/session_data', '/tmp/session_summaries']:
+    os.makedirs(directory, exist_ok=True)
+
+# Hugging Face Spaces specific settings
+SPACES_CONFIG = {
+    'port': 7860,  # Default port for Hugging Face Spaces
+    'host': '0.0.0.0',
+    'workers': 1,  # Single worker for Hugging Face Spaces
+    'timeout': 120,  # Increased timeout for model loading
+    'log_level': 'info'
+}
+
+# Model settings optimized for Hugging Face Spaces
+MODEL_CONFIG = {
+    'model_name': 'meta-llama/Llama-3.2-3B-Instruct',
+    'peft_model_path': 'nada013/mental-health-chatbot',
+    'use_4bit': True,
+    'device': 'cpu'  # Force CPU for Hugging Face Spaces
+} 

requirements.txt

@@ -1,26 +1,27 @@
-transformers==4.49.0
-torch==2.2.0
-sentence-transformers==3.4.1
-langchain==0.3.21
-langchain-community==0.3.20
-langchain-core==0.3.47
-langchain-huggingface==0.1.2
-pydantic==2.10.6
-pydantic-settings==2.8.1
-fastapi==0.115.11
-uvicorn==0.34.0
-python-dotenv==1.0.1
-pytest==7.4.0
-gunicorn==21.2.0
-accelerate==1.5.2
-bitsandbytes==0.45.3
-chromadb==0.6.3
-datasets==3.4.1
-faiss-cpu==1.10.0
-huggingface-hub==0.29.3
-peft==0.15.1
-safetensors==0.5.3
-tokenizers==0.21.1
-tiktoken==0.9.0
-starlette==0.46.1
-websockets==15.0.1 
+transformers>=4.49.0
+torch>=2.2.0
+sentence-transformers>=3.4.1
+langchain>=0.3.21
+langchain-community>=0.3.20
+langchain-core>=0.3.47
+langchain-huggingface>=0.1.2
+pydantic>=2.10.6
+pydantic-settings>=2.8.1
+fastapi>=0.115.11
+uvicorn>=0.34.0
+python-dotenv>=1.0.1
+pytest>=7.4.0
+gunicorn>=21.2.0
+accelerate>=1.5.2
+bitsandbytes>=0.45.3
+chromadb>=0.6.3
+datasets>=3.4.1
+faiss-cpu>=1.10.0
+huggingface-hub>=0.29.3
+peft>=0.15.1
+safetensors>=0.5.3
+tokenizers>=0.21.1
+tiktoken>=0.9.0
+starlette>=0.46.1
+websockets>=15.0.1
+python-multipart>=0.0.6