Update app.py

app.py

@@ -3,10 +3,7 @@ from PIL import Image
 from transformers import AutoProcessor, Blip2ForConditionalGeneration
 import os
 processor = AutoProcessor.from_pretrained("shadowlilac/visor-v2", token=os.getenv('HUGGING_FACE_HUB_TOKEN'))
-model = Blip2ForConditionalGeneration.from_pretrained("shadowlilac/visor-v2", token=os.getenv('HUGGING_FACE_HUB_TOKEN'))
-model.config.vision_config.image_size=768
-model.config.vision_config.patch_size=28
-model.config.vision_config.projection_dim=768
+model = Blip2ForConditionalGeneration.from_pretrained("shadowlilac/visor-v2", ignore_mismatched_sizes=True, token=os.getenv('HUGGING_FACE_HUB_TOKEN'))
 
 def generate_caption(image):
     raw_image = Image.fromarray(image)