Simplify to use only BioMistral base model, remove all adapter references

app.py

@@ -19,11 +19,11 @@ def force_gpu_detection():
 
 class HFMedicalChatbot:
     def __init__(self):
-        # BioMistral model configuration
-        self.model_id = "BioMistral/BioMistral-7B"  # Direct BioMistral model
+        # BioMistral base model configuration
+        BASE = "BioMistral/BioMistral-7B"
         
         print("🔄 Initializing HF Medical Chatbot")
-        print(f"📦 Loading model: {self.model_id}")
+        print(f"📦 Loading base model: {BASE}")
         
         # Force GPU detection first
         try:
@@ -33,12 +33,12 @@ class HFMedicalChatbot:
             print(f"⚠️ GPU detection error: {e}")
             gpu_available = torch.cuda.is_available()
         
-        self._load_model(gpu_available)
+        self._load_model(BASE, gpu_available)
     
-    def _load_model(self, gpu_available):
-        """Load the BioMistral medical model directly"""
+    def _load_model(self, model_id, gpu_available):
+        """Load the BioMistral base model"""
         try:
-            print("🔄 Loading BioMistral model...")
+            print("🔄 Loading BioMistral base model...")
             
             # Determine device strategy
             if gpu_available and torch.cuda.is_available():
@@ -51,16 +51,16 @@ class HFMedicalChatbot:
                 print("💻 Loading BioMistral model on CPU...")
             
             # Load tokenizer
-            print(f"📝 Loading tokenizer: {self.model_id}")
+            print(f"📝 Loading tokenizer: {model_id}")
             self.tokenizer = AutoTokenizer.from_pretrained(
-                self.model_id,
+                model_id,
                 trust_remote_code=True
             )
             
             # Load the model
-            print(f"📦 Loading model: {self.model_id}")
+            print(f"📦 Loading model: {model_id}")
             self.model = MistralForCausalLM.from_pretrained(
-                self.model_id,
+                model_id,
                 trust_remote_code=True,
                 device_map="auto",
                 torch_dtype=dtype,
@@ -71,7 +71,7 @@ class HFMedicalChatbot:
             if self.tokenizer.pad_token is None:
                 self.tokenizer.pad_token = self.tokenizer.eos_token
             
-            print(f"✅ BioMistral model loaded successfully on {device.upper()}!")
+            print(f"✅ BioMistral base model loaded successfully on {device.upper()}!")
             
         except Exception as e:
             print(f"❌ Error loading BioMistral model: {e}")