Update app.py

app.py

@@ -16,10 +16,13 @@ bnb_config = BitsAndBytesConfig(
         )
 
 # Load models and tokenizer efficiently
-config = PeftConfig.from_pretrained("phearion/bigbrain-v0.0.1")
+config = PeftConfig.from_pretrained(model_id)
 tokenizer = AutoTokenizer.from_pretrained(config.base_model_name_or_path)
 model = AutoModelForCausalLM.from_pretrained(config.base_model_name_or_path, quantization_config=bnb_config)
 
+# Load the Lora model
+model = PeftModel.from_pretrained(model, model_id)
+
 def greet(text):
     with torch.no_grad():  # Disable gradient calculation for inference
         batch = tokenizer(f'"{text}" ->:', return_tensors='pt')  # Move tensors to device