chore: auth and logging

app.py

@@ -2,13 +2,26 @@ import gradio as gr
 import torch
 from transformers import AutoTokenizer, AutoModelForCausalLM, BitsAndBytesConfig
 import re
+import logging
+import os
+
+# Configure logging
+logging.basicConfig(
+    level=logging.INFO,
+    format='%(asctime)s - %(name)s - %(levelname)s - %(message)s',
+    handlers=[
+        logging.StreamHandler(),
+        logging.FileHandler('medgemma_app.log')
+    ]
+)
+logger = logging.getLogger(__name__)
 
 class MedGemmaSymptomAnalyzer:
     def __init__(self):
         self.model = None
         self.tokenizer = None
         self.model_loaded = False
-        print("Initializing MedGemma Symptom Analyzer...")
+        logger.info("Initializing MedGemma Symptom Analyzer...")
     
     def load_model(self):
         """Load MedGemma model with optimizations for deployment"""
@@ -16,32 +29,43 @@ class MedGemmaSymptomAnalyzer:
             return True
             
         model_name = "google/medgemma-4b-it"
-        print(f"Loading model: {model_name}")
+        logger.info(f"Loading model: {model_name}")
         
         try:
+            # Get HF token from environment (set in Hugging Face Spaces secrets)
+            hf_token = os.getenv("HF_TOKEN")
+            if hf_token:
+                logger.info("Using HF_TOKEN for authentication")
+            else:
+                logger.warning("HF_TOKEN not found in environment variables")
+            
             # First try without quantization for CPU compatibility
-            print("Loading tokenizer...")
-            self.tokenizer = AutoTokenizer.from_pretrained(model_name)
+            logger.info("Loading tokenizer...")
+            self.tokenizer = AutoTokenizer.from_pretrained(
+                model_name,
+                token=hf_token
+            )
             
-            print("Loading model...")
+            logger.info("Loading model...")
             # Simplified loading for CPU/compatibility
             self.model = AutoModelForCausalLM.from_pretrained(
                 model_name,
                 torch_dtype=torch.float32,  # Use float32 for CPU
                 device_map="cpu",  # Force CPU for compatibility
-                low_cpu_mem_usage=True
+                low_cpu_mem_usage=True,
+                token=hf_token
             )
             
             if self.tokenizer.pad_token is None:
                 self.tokenizer.pad_token = self.tokenizer.eos_token
                 
             self.model_loaded = True
-            print("Model loaded successfully!")
+            logger.info("Model loaded successfully!")
             return True
                 
         except Exception as e:
-            print(f"Error loading model: {e}")
-            print("Falling back to demo mode...")
+            logger.error(f"Failed to load model {model_name}: {str(e)}", exc_info=True)
+            logger.warning("Falling back to demo mode due to model loading failure")
             self.model = None
             self.tokenizer = None
             self.model_loaded = False