Upload folder using huggingface_hub

app.py

@@ -1,7 +1,6 @@
 import shutil
 import gradio as gr
 import torch
-from fastapi import FastAPI
 import os
 import tempfile
 from Infer import Infer
@@ -37,9 +36,9 @@ def generate(video, textbox_in, candbox_in):
     video = video if video else "none"
 
     text_en_out = handler.generate(textbox_in, eval(candbox_in), video)
-    print(text_en_out)
     textbox_out = text_en_out
     #torch.cuda.empty_cache()
+    print(textbox_out)
     return textbox_out
 
 
@@ -52,9 +51,10 @@ if not os.path.exists("temp"):
 #print(torch.cuda.memory_allocated())
 #print(torch.cuda.max_memory_allocated())
 
+
+video = gr.Video(label="Input Video")
 question_box = gr.Textbox(
-    show_label=False, placeholder="Enter question", container=False
-    )
+show_label=False, placeholder="Enter question", container=False)
 
 candidates_box = gr.Textbox(
     show_label=False, placeholder="Enter a list of options", container=False
@@ -66,45 +66,74 @@ with gr.Blocks(title='T-MoENet', theme=gr.themes.Default(), css=block_css) as de
     state_ = gr.State()
     first_run = gr.State()
     images_tensor = gr.State()
-
-    with gr.Row():
+    cur_dir = os.path.dirname(os.path.abspath(__file__))
+    
+    with gr.Column():
+        with gr.Column(scale=3):
+            gr.Interface(
+                generate,
+                [video, question_box, candidates_box],
+                ["text"]
+            )   
         with gr.Column(scale=3):
-            video = gr.Video(label="Input Video")
-            cur_dir = os.path.dirname(os.path.abspath(__file__))
-            print(cur_dir)
             gr.Examples(
-                examples=[
-                    [
-                        cur_dir + "/videos/3249402410.mp4",
-                        "What did the lady in black on the left do after she finished spreading the sauce on her pizza?",
-                            "['slice the pizza', 'cut the meat', 'point', 'put cheese', 'put on plate']"
+                    examples=[
+                        [
+                            cur_dir + "/videos/3249402410.mp4",
+                            "What did the lady in black on the left do after she finished spreading the sauce on her pizza?",
+                                "['slice the pizza', 'cut the meat', 'point', 'put cheese', 'put on plate']"
+                        ],
+                        [
+                            cur_dir + "/videos/4882821564.mp4",
+                            "Why did the boy clap his hands when he ran to the christmas tree?",
+                            "['adjust the tree', 'get away the dust', 'dancing', 'pressed a button to activate', 'presents']"
+                        ],
+                        [
+                            cur_dir + "/videos/6233408665.mp4",
+                            "What did the people on the sofa do after the lady in pink finished singing?",
+                            "['sitting', 'give it to the girl', 'take music sheet', 'clap', 'walk in circles']"
+                        ],
                     ],
-                    [
-                        cur_dir + "/videos/4882821564.mp4",
-                        "Why did the boy clap his hands when he ran to the christmas tree?",
-                        "['adjust the tree', 'get away the dust', 'dancing', 'pressed a button to activate', 'presents']"
-                    ],
-                    [
-                        cur_dir + "/videos/6233408665.mp4",
-                        "What did the people on the sofa do after the lady in pink finished singing?",
-                        "['sitting', 'give it to the girl', 'take music sheet', 'clap', 'walk in circles']"
-                    ],
-                ],
-                inputs=[video, question_box, candidates_box],
-            )
-
-        with gr.Column(scale=3):
-            chatbot = gr.Chatbot(label="T-MoENet", bubble_full_width=True)
-            with gr.Row():
-                with gr.Column(scale=4):
-                    question_box.render()
-                with gr.Column(scale=4): 
-                    candidates_box.render()
-                with gr.Column(scale=1, min_width=50):
-                    submit_btn = gr.Button(
-                        value="Send", variant="primary", interactive=True
-                    )
-
-    submit_btn.click(generate, [video, question_box, candidates_box], [chatbot])
+                    inputs=[video, question_box, candidates_box]
+                ) 
+
+    # with gr.Row():
+    #     with gr.Column(scale=3):
+    #         video = gr.Video(label="Input Video")
+    #         cur_dir = os.path.dirname(os.path.abspath(__file__))
+    #         print(cur_dir)
+    #         gr.Examples(
+    #             examples=[
+    #                 [
+    #                     cur_dir + "/videos/3249402410.mp4",
+    #                     "What did the lady in black on the left do after she finished spreading the sauce on her pizza?",
+    #                         "['slice the pizza', 'cut the meat', 'point', 'put cheese', 'put on plate']"
+    #                 ],
+    #                 [
+    #                     cur_dir + "/videos/4882821564.mp4",
+    #                     "Why did the boy clap his hands when he ran to the christmas tree?",
+    #                     "['adjust the tree', 'get away the dust', 'dancing', 'pressed a button to activate', 'presents']"
+    #                 ],
+    #                 [
+    #                     cur_dir + "/videos/6233408665.mp4",
+    #                     "What did the people on the sofa do after the lady in pink finished singing?",
+    #                     "['sitting', 'give it to the girl', 'take music sheet', 'clap', 'walk in circles']"
+    #                 ],
+    #             ],
+    #             inputs=[video, question_box, candidates_box],
+    #         )
+
+        # with gr.Column(scale=3):
+            
+        #     with gr.Row():
+        #         with gr.Column(scale=4):
+        #             question_box.render()
+        #         with gr.Column(scale=4): 
+        #             candidates_box.render()
+        #         with gr.Column(scale=1, min_width=50):
+        #             submit_btn = gr.Button(
+        #                 value="Send", variant="primary", interactive=True
+        #             )
+    #submit_btn.click(generate, [video, question_box, candidates_box], [chatbot])
 
 demo.launch(share=True)

requirements.txt

@@ -42,7 +42,6 @@ matplotlib==3.7.5
 mdurl==0.1.2
 mpmath==1.3.0
 networkx==3.1
-numpy==1.24.4
 nvidia-cublas-cu11==11.10.3.66
 nvidia-cuda-cupti-cu11==11.7.101
 nvidia-cuda-nvrtc-cu11==11.7.99
@@ -57,7 +56,7 @@ nvidia-nvtx-cu11==11.7.91
 opencv-python==4.10.0.84
 orjson==3.10.6
 packaging==24.1
-pandas==2.0.3
+pandas
 pillow==10.4.0
 pkgutil_resolve_name==1.3.10
 pydantic==2.8.2