Update app.py

app.py

@@ -4,18 +4,17 @@ import gradio as gr
 device = 'cuda' if torch.cuda.is_available() else 'cpu'
 processor=AutoProcessor.from_pretrained("alibidaran/General_image_captioning")
 model=AutoModelForCausalLM.from_pretrained("alibidaran/General_image_captioning").to(device)
-def generate_caption(image,length):
+def generate_caption(image):
   encoded=processor(images=image, return_tensors="pt").to(device)
   pixels=encoded['pixel_values'].to(device)
   with torch.no_grad():
-    generated_ids=model.generate(pixel_values=pixels,max_length=length)
+    generated_ids=model.generate(pixel_values=pixels,max_length=10)
   generated_caption = processor.batch_decode(generated_ids, skip_special_tokens=True)[0]
   return generated_caption 
 demo=gr.Interface(
     fn=generate_caption,
     inputs=[
-        gr.Image(type='pil',flagging_options=["blurry", "incorrect", "other"]),
-        gr.Slider(10,50,value=10)
+        gr.Image(type='pil'),
     ],
     outputs= 'label',
     examples=['sample.jpg','sample1.jpg','sample2.jpg'],