Update app.py

app.py

@@ -1,3 +1,36 @@
 import gradio as gr
+from huggingface_hub import InferenceApi
+import requests
+from PIL import Image
+import io
 
-gr.load("models/Salesforce/blip-image-captioning-large").launch()
+# Initialize the Hugging Face Inference API
+model_id = "Salesforce/blip-image-captioning-large"
+api = InferenceApi(repo_id=model_id)
+
+def caption_image(image):
+    # Convert the PIL image to bytes
+    buffered = io.BytesIO()
+    image.save(buffered, format="JPEG")
+    img_bytes = buffered.getvalue()
+    
+    # Call the Hugging Face inference API
+    response = api(inputs=img_bytes)
+    
+    # Check the response and format it properly
+    if isinstance(response, list) and response:
+        return response[0]["generated_text"]
+    else:
+        return "Error generating caption"
+
+# Set up the Gradio interface
+interface = gr.Interface(
+    fn=caption_image,
+    inputs=gr.inputs.Image(type="pil"),
+    outputs="text",
+    title="Image Captioning",
+    description="Generate captions for images using the BLIP model."
+)
+
+# Launch the interface
+interface.launch()