fix

app.py

@@ -10,7 +10,7 @@ model, _, transform = open_clip.create_model_and_transforms(
 def output_generate(image):
     im = transform(image).unsqueeze(0)
     generated = model.generate(im)
-    return open_clip.decode(generated).split("<end_of_text>")[0].replace("<start_of_text>", "")
+    return open_clip.decode(generated[0]).split("<end_of_text>")[0].replace("<start_of_text>", "")
 
-iface = gr.Interface(fn=output_generate, inputs=gr.Image(type="pil"), outputs="text")
+iface = gr.Interface(fn=output_generate, inputs=gr.Ifimage(type="pil"), outputs="text")
 iface.launch()