Update app.py

app.py

@@ -1,7 +1,5 @@
 from PIL import Image
-import requests
 import gradio as gr
-
 from transformers import BlipProcessor, BlipForConditionalGeneration
 
 model_id = "Salesforce/blip-image-captioning-base"
@@ -9,11 +7,21 @@ model_id = "Salesforce/blip-image-captioning-base"
 model = BlipForConditionalGeneration.from_pretrained(model_id)
 processor = BlipProcessor.from_pretrained(model_id)
 
-def launch(input):
-    image = Image.open(requests.get(input, stream=True).raw).convert('RGB')
+def generate_caption(image_path):
+    # Load the image directly from the path
+    image = Image.open(image_path).convert('RGB')
+    
+    # Process the image to generate tensor inputs
     inputs = processor(image, return_tensors="pt")
+    
+    # Generate caption for the image
     out = model.generate(**inputs)
+    
+    # Decode and return the generated caption
     return processor.decode(out[0], skip_special_tokens=True)
 
-iface = gr.Interface(launch, inputs="text", outputs="text")
-iface.launch()
+# Gradio interface setup to accept image input and produce text output
+iface = gr.Interface(generate_caption, inputs="image", outputs="text")
+
+# Launch the interface
+iface.launch()