Spaces:
Runtime error
Runtime error
Commit
Β·
acd4f5a
1
Parent(s):
2296892
Update app.py
Browse files
app.py
CHANGED
|
@@ -113,7 +113,7 @@ def generate(image1, video, textbox_in, first_run, state, state_, images_tensor)
|
|
| 113 |
if flag:
|
| 114 |
state.append_message(state.roles[0], textbox_in + "\n" + show_images)
|
| 115 |
state.append_message(state.roles[1], textbox_out)
|
| 116 |
-
|
| 117 |
return (state, state_, state.to_gradio_chatbot(), False, gr.update(value=None, interactive=True), images_tensor, gr.update(value=image1 if os.path.exists(image1) else None, interactive=True), gr.update(value=video if os.path.exists(video) else None, interactive=True))
|
| 118 |
|
| 119 |
def regenerate(state, state_):
|
|
@@ -232,18 +232,82 @@ with gr.Blocks(title='Video-LLaVAπ', theme=gr.themes.Default(), css=block_css
|
|
| 232 |
f"{cur_dir}/examples/sample_demo_1.mp4",
|
| 233 |
"Why is this video funny?",
|
| 234 |
],
|
|
|
|
|
|
|
|
|
|
|
|
|
| 235 |
[
|
| 236 |
f"{cur_dir}/examples/sample_demo_3.mp4",
|
| 237 |
-
"Can you identify any safety hazards in this video?"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 238 |
],
|
| 239 |
[
|
| 240 |
-
f"{cur_dir}/examples/
|
| 241 |
"Describe the video.",
|
| 242 |
],
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 243 |
[
|
| 244 |
f"{cur_dir}/examples/sample_demo_22.mp4",
|
| 245 |
"Describe the activity in the video.",
|
| 246 |
],
|
|
|
|
|
|
|
|
|
|
|
|
|
| 247 |
],
|
| 248 |
inputs=[video, textbox],
|
| 249 |
)
|
|
|
|
| 113 |
if flag:
|
| 114 |
state.append_message(state.roles[0], textbox_in + "\n" + show_images)
|
| 115 |
state.append_message(state.roles[1], textbox_out)
|
| 116 |
+
torch.cuda.empty_cache()
|
| 117 |
return (state, state_, state.to_gradio_chatbot(), False, gr.update(value=None, interactive=True), images_tensor, gr.update(value=image1 if os.path.exists(image1) else None, interactive=True), gr.update(value=video if os.path.exists(video) else None, interactive=True))
|
| 118 |
|
| 119 |
def regenerate(state, state_):
|
|
|
|
| 232 |
f"{cur_dir}/examples/sample_demo_1.mp4",
|
| 233 |
"Why is this video funny?",
|
| 234 |
],
|
| 235 |
+
[
|
| 236 |
+
f"{cur_dir}/examples/sample_demo_2.mp4",
|
| 237 |
+
"what is strange about this video?",
|
| 238 |
+
],
|
| 239 |
[
|
| 240 |
f"{cur_dir}/examples/sample_demo_3.mp4",
|
| 241 |
+
"Can you identify any safety hazards in this video?",
|
| 242 |
+
],
|
| 243 |
+
[
|
| 244 |
+
f"{cur_dir}/examples/sample_demo_4.mp4",
|
| 245 |
+
"what is unusual about this video?",
|
| 246 |
+
],
|
| 247 |
+
[
|
| 248 |
+
f"{cur_dir}/examples/sample_demo_6.mp4",
|
| 249 |
+
"Can you write a short poem inspired fom the video.",
|
| 250 |
+
],
|
| 251 |
+
[
|
| 252 |
+
f"{cur_dir}/examples/sample_demo_7.mp4",
|
| 253 |
+
"Create a short fairy tale with a moral lesson inspired by the video.",
|
| 254 |
+
],
|
| 255 |
+
[
|
| 256 |
+
f"{cur_dir}/examples/sample_demo_8.mp4",
|
| 257 |
+
"Where is this video taken from? What place/landmark is shown in the video?",
|
| 258 |
+
],
|
| 259 |
+
[
|
| 260 |
+
f"{cur_dir}/examples/sample_demo_10.mp4",
|
| 261 |
+
"Describe the video in detail.",
|
| 262 |
+
],
|
| 263 |
+
[
|
| 264 |
+
f"{cur_dir}/examples/sample_demo_11.mp4",
|
| 265 |
+
"What is the main focus of the video.",
|
| 266 |
+
],
|
| 267 |
+
[
|
| 268 |
+
f"{cur_dir}/examples/sample_demo_12.mp4",
|
| 269 |
+
"What does the woman use to split the logs and how does she do it?",
|
| 270 |
],
|
| 271 |
[
|
| 272 |
+
f"{cur_dir}/examples/sample_demo_13.mp4",
|
| 273 |
"Describe the video.",
|
| 274 |
],
|
| 275 |
+
[
|
| 276 |
+
f"{cur_dir}/examples/sample_demo_14.mp4",
|
| 277 |
+
"What is the video about?",
|
| 278 |
+
],
|
| 279 |
+
[
|
| 280 |
+
f"{cur_dir}/examples/sample_demo_15.mp4",
|
| 281 |
+
"What is the main challenge faced by the people on the boat?",
|
| 282 |
+
],
|
| 283 |
+
[
|
| 284 |
+
f"{cur_dir}/examples/sample_demo_17.mp4",
|
| 285 |
+
"What activities are the woman and the dog engaged in the video?",
|
| 286 |
+
],
|
| 287 |
+
[
|
| 288 |
+
f"{cur_dir}/examples/sample_demo_18.mp4",
|
| 289 |
+
"Describe the video in detail.",
|
| 290 |
+
],
|
| 291 |
+
[
|
| 292 |
+
f"{cur_dir}/examples/sample_demo_19.mp4",
|
| 293 |
+
"What is the video about?",
|
| 294 |
+
],
|
| 295 |
+
[
|
| 296 |
+
f"{cur_dir}/examples/sample_demo_20.mp4",
|
| 297 |
+
"Can you describe the process for cleaning a screen window as demonstrated in the video?",
|
| 298 |
+
],
|
| 299 |
+
[
|
| 300 |
+
f"{cur_dir}/examples/sample_demo_21.mp4",
|
| 301 |
+
"Step-by-step describe the cooking recipe shown in the video.",
|
| 302 |
+
],
|
| 303 |
[
|
| 304 |
f"{cur_dir}/examples/sample_demo_22.mp4",
|
| 305 |
"Describe the activity in the video.",
|
| 306 |
],
|
| 307 |
+
[
|
| 308 |
+
f"{cur_dir}/examples/sample_demo_23.mp4",
|
| 309 |
+
"Briefly describe the video.",
|
| 310 |
+
],
|
| 311 |
],
|
| 312 |
inputs=[video, textbox],
|
| 313 |
)
|