Update app.py

app.py

@@ -92,8 +92,8 @@ def generate_caption_and_image(image, f, p, d):
         num_inference_steps=50,
         max_sequence_length=512,
         generator=torch.Generator("cpu").manual_seed(0)
-        ).images[0]
-        return image
+        ).images
+        return image[0], image[1]