Update app.py

app.py

@@ -146,49 +146,55 @@ def inference_image(img):
     return {imagenet_id_to_classname[str(i)]: float(prediction[i]) for i in range(1000)}
 
 
-demo = gr.Blocks()
-with demo:
-    gr.Markdown(
-        """
-        # VideoMamba-Ti
-        Gradio demo for <a href='https://github.com/OpenGVLab/VideoMamba' target='_blank'>VideoMamba</a>: To use it, simply upload your video, or click one of the examples to load them. Read more at the links below.
-        """
-    )
-
-    # with gr.Tab("Video"):
-    #     # with gr.Box():
-    with gr.Row():
-        with gr.Column():
-            with gr.Row():
-                input_video = gr.Video(label='Input Video', height=360)
-                # input_video = load_video(input_video)
-            with gr.Row():
-                submit_video_button = gr.Button('Submit')
-        with gr.Column():
-                label_video = gr.Label(num_top_classes=5)
-        # with gr.Row():
-        #     gr.Examples(examples=['./videos/hitting_baseball.mp4', './videos/hoverboarding.mp4', './videos/yoga.mp4'], inputs=input_video, outputs=label_video, fn=inference_video, cache_examples=True)
-
-    # with gr.Tab("Image"):
-    #     # with gr.Box():
-    #     with gr.Row():
-    #         with gr.Column():
-    #             with gr.Row():
-    #                 input_image = gr.Image(label='Input Image', type='pil', height=360)
-    #             with gr.Row():
-    #                 submit_image_button = gr.Button('Submit')
-    #         with gr.Column():
-    #             label_image = gr.Label(num_top_classes=5)
-        # with gr.Row():
-        #     gr.Examples(examples=['./images/cat.png', './images/dog.png', './images/panda.png'], inputs=input_image, outputs=label_image, fn=inference_image, cache_examples=True)
-
-    gr.Markdown(
-        """
-        <p style='text-align: center'><a href='https://arxiv.org/abs/2403.06977' target='_blank'>VideoMamba: State Space Model for Efficient Video Understanding</a> | <a href='https://github.com/OpenGVLab/VideoMamba' target='_blank'>Github Repo</a></p>
-        """
-    )
-
-    submit_video_button.click(fn=ultra_inference_video, inputs=input_video, outputs=label_video)
-    # submit_image_button.click(fn=inference_image, inputs=input_image, outputs=label_image)
-
-demo.queue(max_size=20).launch()
+demo = gr.Interface(
+    fn = ultra_inference_video,
+    inputs = gr.Video()
+    outputs = "label"
+           
+)
+    # demo = gr.Blocks()
+# with demo:
+    # # gr.Markdown(
+    # #     """
+    # #     # VideoMamba-Ti
+    # #     Gradio demo for <a href='https://github.com/OpenGVLab/VideoMamba' target='_blank'>VideoMamba</a>: To use it, simply upload your video, or click one of the examples to load them. Read more at the links below.
+    # #     """
+    # # )
+
+    # # with gr.Tab("Video"):
+    # #     # with gr.Box():
+    # with gr.Row():
+    #     with gr.Column():
+    #         with gr.Row():
+    #             input_video = gr.Video(label='Input Video', height=360)
+    #             # input_video = load_video(input_video)
+    #         with gr.Row():
+    #             submit_video_button = gr.Button('Submit')
+    #     with gr.Column():
+    #             label_video = gr.Label(num_top_classes=5)
+    #     # with gr.Row():
+    #     #     gr.Examples(examples=['./videos/hitting_baseball.mp4', './videos/hoverboarding.mp4', './videos/yoga.mp4'], inputs=input_video, outputs=label_video, fn=inference_video, cache_examples=True)
+
+    # # with gr.Tab("Image"):
+    # #     # with gr.Box():
+    # #     with gr.Row():
+    # #         with gr.Column():
+    # #             with gr.Row():
+    # #                 input_image = gr.Image(label='Input Image', type='pil', height=360)
+    # #             with gr.Row():
+    # #                 submit_image_button = gr.Button('Submit')
+    # #         with gr.Column():
+    # #             label_image = gr.Label(num_top_classes=5)
+    #     # with gr.Row():
+    #     #     gr.Examples(examples=['./images/cat.png', './images/dog.png', './images/panda.png'], inputs=input_image, outputs=label_image, fn=inference_image, cache_examples=True)
+
+    # gr.Markdown(
+    #     """
+    #     <p style='text-align: center'><a href='https://arxiv.org/abs/2403.06977' target='_blank'>VideoMamba: State Space Model for Efficient Video Understanding</a> | <a href='https://github.com/OpenGVLab/VideoMamba' target='_blank'>Github Repo</a></p>
+    #     """
+    # )
+
+    # submit_video_button.click(fn=ultra_inference_video, inputs=input_video, outputs=label_video)
+    # # submit_image_button.click(fn=inference_image, inputs=input_image, outputs=label_image)
+
+demo.launch()