Update app.py

app.py

@@ -12,21 +12,47 @@ from typing import Dict, List
 import sqlite3
 from contextlib import contextmanager
 import re
-import os
 from werkzeug.utils import secure_filename
 
 app = Flask(__name__)
 
-UPLOAD_FOLDER = 'uploads'
+PORT = int(os.environ.get("PORT", 7860))
+
+UPLOAD_FOLDER = '/tmp/uploads'  # Change to tmp directory for Spaces
 ALLOWED_EXTENSIONS = {'py'}
 app.config['UPLOAD_FOLDER'] = UPLOAD_FOLDER
 os.makedirs(UPLOAD_FOLDER, exist_ok=True)
 # Database configuration
-DATABASE_PATH = 'chat_database.db'
+DATABASE_PATH = '/tmp/chat_database.db'
 
 # Initialize LangChain with Ollama LLM
-llm = Ollama(model="mistral:7b")
+from transformers import AutoTokenizer, AutoModelForCausalLM
+import torch
+
+# Load model and tokenizer
+model_name = "mistralai/Mistral-7B-Instruct-v0.1"
+tokenizer = AutoTokenizer.from_pretrained(model_name)
+model = AutoModelForCausalLM.from_pretrained(model_name)
+
+class HuggingFaceLLM:
+    def __init__(self, model, tokenizer):
+        self.model = model
+        self.tokenizer = tokenizer
+    
+    def predict(self, prompt):
+        inputs = self.tokenizer(prompt, return_tensors="pt", max_length=2048, truncation=True)
+        with torch.no_grad():
+            outputs = self.model.generate(
+                inputs["input_ids"],
+                max_length=2048,
+                num_return_sequences=1,
+                temperature=0.7,
+                pad_token_id=self.tokenizer.eos_token_id
+            )
+        response = self.tokenizer.decode(outputs[0], skip_special_tokens=True)
+        return response
 
+llm = HuggingFaceLLM(model, tokenizer)
 
 @contextmanager
 def get_db_connection():
@@ -606,4 +632,5 @@ def home():
 
 
 if __name__ == "__main__":
-    app.run(debug=True)
+    init_db()
+    app.run(host="0.0.0.0", port=PORT)