Update app.py

app.py

@@ -122,15 +122,15 @@ model = Phi3WithImage(phi3_model_name,  image_encoder, image_embed_dim, bnb_conf
 # Load LoRA model
 model.phi3 = PeftModel.from_pretrained(model.phi3, lora_model_path, device_map="auto", offload_dir='./offload')
 model.phi3 = model.phi3.merge_and_unload()
-model.phi3.save_pretrained("merged_model_fp16")
+#model.phi3.save_pretrained("merged_model_fp16")
 
 #model.phi3 = PeftModel.from_pretrained(model.phi3, lora_model_path, device_map="auto")
-model.phi3 = AutoModelForCausalLM.from_pretrained(
-    "merged_model_fp16",
-    quantization_config=bnb_config,
-    torch_dtype=torch.bfloat16,
-    device_map="auto"
-)
+#model.phi3 = AutoModelForCausalLM.from_pretrained(
+#    "merged_model_fp16",
+#    quantization_config=bnb_config,
+#    torch_dtype=torch.bfloat16,
+#    device_map="auto"
+#)
 model.eval() # Set to evaluation mode
 
 # 3. Inference Function