lixin4ever ZYX123 commited on
Commit
a5deedf
1 Parent(s): e52682b

add requirements.txt (#3)

Browse files

- add requirements.txt (e818756da384130f584390d4c7707ecf72384d29)
- add requirements.txt (0658a9d415b8cd677486731ecac8757a4d15f2a3)


Co-authored-by: yongxinzhu <zyx123@users.noreply.huggingface.co>

Files changed (2) hide show
  1. app.py +2 -2
  2. requirements.txt +41 -0
app.py CHANGED
@@ -321,11 +321,11 @@ with gr.Blocks(title='VideoLLaMA 2 🔥🚀🔥', theme=theme, css=block_css) as
321
  examples=[
322
  [
323
  f"{cur_dir}/examples/00000368.mp4",
324
- "Where is the loudest instrument?",
325
  ],
326
  [
327
  f"{cur_dir}/examples/00003491.mp4",
328
- "Is the instrument on the left louder than the instrument on the right?",
329
  ],
330
  ],
331
  inputs=[video, textbox],
 
321
  examples=[
322
  [
323
  f"{cur_dir}/examples/00000368.mp4",
324
+ "Please describe the video with audio information:",
325
  ],
326
  [
327
  f"{cur_dir}/examples/00003491.mp4",
328
+ "Where is the loudest instrument?",
329
  ],
330
  ],
331
  inputs=[video, textbox],
requirements.txt ADDED
@@ -0,0 +1,41 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # basic dependencies
2
+ torch==2.2.0
3
+ torchaudio==2.2.0
4
+ torchvision==0.17.0
5
+ transformers==4.42.3
6
+ tokenizers==0.19.1
7
+ deepspeed==0.13.1
8
+ accelerate==0.26.1
9
+ peft==0.4.0
10
+ timm==1.0.3
11
+ numpy==1.24.4
12
+ # data processing
13
+ decord==0.6.0
14
+ imageio==2.34.0
15
+ imageio-ffmpeg==0.4.9
16
+ moviepy==1.0.3
17
+ scenedetect==0.6.3
18
+ opencv-python==4.6.0.66
19
+ pysubs2
20
+ librosa
21
+ pytorchvideo
22
+ # misc
23
+ scikit-learn==1.2.2
24
+ huggingface_hub==0.23.4
25
+ sentencepiece==0.1.99
26
+ shortuuid
27
+ einops==0.6.1
28
+ einops-exts==0.0.4
29
+ bitsandbytes==0.43.0
30
+ pydantic>=2.0
31
+ markdown2[all]
32
+ gradio==3.50.0
33
+ gradio_client==0.6.1
34
+ httpx==0.24.1
35
+ openai==1.33.0
36
+ requests
37
+ uvicorn
38
+ fastapi
39
+ tensorboard
40
+ wandb
41
+ tabulate