Update main.py

main.py

@@ -8,26 +8,17 @@ from qwen_vl_utils import process_vision_info
 
 # Specify the model path or identifier.
 MODEL_PATH = "Ananthu01/qwen2.5_vl_finetuned_model"
-# MODEL_PATH = "Qwen/Qwen2.5-VL-3B-Instruct"
 
 # Initialize the Qwen2.5 VL model.
-# model = AutoModel.from_pretrained(
-#     MODEL_PATH,
-#     trust_remote_code=True,
-#     device_map="cpu"  # Ensures the model is loaded on CPU.
-# )
-
 model = Qwen2_5_VLForConditionalGeneration.from_pretrained(
     MODEL_PATH,
     trust_remote_code=True,
     device_map="cpu",  # Ensures the model is loaded on CPU.
-    wbits=4,          # Example parameter for 4-bit quantization
-    groupsize=128,     # Example parameter for groupsize setting
     use_safetensors=True
 )
 
 # Load the processor.
-processor = AutoProcessor.from_pretrained(MODEL_PATH, trust_remote_code=True)
+processor = AutoProcessor.from_pretrained("Qwen/Qwen2.5-VL-3B-Instruct", trust_remote_code=True)
 
 # Load the tokenizer.
 tokenizer = AutoTokenizer.from_pretrained(MODEL_PATH, trust_remote_code=True)
@@ -124,3 +115,4 @@ async def generate_output(image_file: UploadFile = File(...)):
 
     # Return the generated text.
     return generated_text
+