Create app.py

app.py

@@ -0,0 +1,25 @@
+import gradio as gr
+from transformers import BlipProcessor, BlipForConditionalGeneration
+import torch
+from PIL import Image
+from io import BytesIO
+import requests
+
+# Load model and processor
+model_name = "Salesforce/blip-image-captioning-large"
+processor = BlipProcessor.from_pretrained(model_name)
+model = BlipForConditionalGeneration.from_pretrained(model_name)
+
+def generate_caption(image):
+    # Preprocess the image
+    inputs = processor(image, return_tensors="pt")
+    # Generate caption
+    with torch.no_grad():
+        outputs = model.generate(**inputs)
+    # Decode and return caption
+    caption = processor.decode(outputs[0], skip_special_tokens=True)
+    return caption
+
+# Create a Gradio interface
+iface = gr.Interface(fn=generate_caption, inputs="image", outputs="text")
+iface.launch(share=True)  # `share=True` to get a public link