commit

app.py

@@ -0,0 +1,25 @@
+import gradio as gr
+from transformers import BlipProcessor, BlipForConditionalGeneration
+
+# Load the model and tokenizer
+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(images=image, return_tensors="pt")
+
+    # Generate caption using the model
+    caption = model.generate(**inputs)
+
+    # Decode the output caption
+    decoded_caption = processor.decode(caption[0], skip_special_tokens=True)
+    return decoded_caption
+
+# Define the Gradio interface
+inputs = gr.inputs.Image(label="Upload an image")
+outputs = gr.outputs.Textbox(label="Generated Caption")
+
+# Create the Gradio app 
+gr.Interface(fn=generate_caption, inputs=inputs, outputs=outputs).launch()