Update app.py

app.py

@@ -1,18 +1,24 @@
 import gradio as gr
 from PIL import Image
-processor = AutoProcessor.from_pretrained(daliavanilla/BLIP-Radiology-model)
-model = BlipForConditionalGeneration.from_pretrained(daliavanilla/BLIP-Radiology-model)
+import torch
+from transformers import BlipForConditionalGeneration, AutoProcessor
+
+# Load processor and model from Hugging Face Hub
+processor = AutoProcessor.from_pretrained("daliavanilla/BLIP-Radiology-model")
+model = BlipForConditionalGeneration.from_pretrained("daliavanilla/BLIP-Radiology-model")
+
+# Use GPU if available
+device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+model.to(device)
 
 # Define the prediction function
 def generate_caption(image):
     # Process the image
     image = Image.fromarray(image)
-    #inputs = tokenizer(image, return_tensors="pt")
-    inputs = processor(images=image, return_tensors="pt")#.to(device)
-    pixel_values = inputs.pixel_values
+    inputs = processor(images=image, return_tensors="pt").to(device)
 
     # Generate caption
-    generated_ids = model.generate(pixel_values=pixel_values, max_length=50)
+    generated_ids = model.generate(pixel_values=inputs.pixel_values, max_length=50)
     generated_caption = processor.batch_decode(generated_ids, skip_special_tokens=True)[0]
 
     return generated_caption
@@ -20,7 +26,7 @@ def generate_caption(image):
 # Define the Gradio interface
 interface = gr.Interface(
     fn=generate_caption,
-    inputs=gr.Image(),
+    inputs=gr.Image(type="numpy"),  # Ensure the image type is correctly handled by PIL
     outputs=gr.Textbox(),
     live=True
 )