File size: 1,351 Bytes
91fb4ef
 
 
 
b613c3c
 
91fb4ef
 
b1855df
91fb4ef
dfc94bc
 
 
 
91fb4ef
 
12bcca7
91fb4ef
 
 
 
 
5d3ad17
0d34ea8
0328e32
0d34ea8
 
5d3ad17
91fb4ef
5d3ad17
 
72e9451
0328e32
72e9451
c5911ab
91fb4ef
 
 
cb66746
 
250d86f
91fb4ef
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
ecd5028
b613c3c
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
numpy>=1.26.4

# to quote a-r-r-o-w/finetrainers:
# It is recommended to use Pytorch 2.5.1 or above for training. Previous versions can lead to completely black videos, OOM errors, or other issues and are not tested.

# on some system (Python 3.13+) those do not work:
torch==2.5.1
torchvision==0.20.1
torchao>=0.7.0

# datasets 3.4.0 replaces decord by torchvision
# let's free it for now
datasets==3.3.2

huggingface_hub
hf_transfer>=0.1.8
diffusers @ git+https://github.com/huggingface/diffusers.git@main
transformers>=4.45.2

accelerate
bitsandbytes
peft>=0.12.0

# For GPU monitoring of NVIDIA chipsets
# you probably won't be able to install that on macOS
# pynvml

# eva-decord is missing get_batch it seems
eva-decord==0.6.1
# decord

finetrainers @ git+https://github.com/a-r-r-o-w/finetrainers.git@main
# temporary fix for pip install bug:
#finetrainers @ git+https://github.com/jbilcke-hf/finetrainers-patches.git@fix_missing_sft_trainer_files

wandb
pandas
sentencepiece>=0.2.0
imageio
imageio-ffmpeg
torchdata==0.11.0

# for youtube video download
pytube
pytubefix

# for scene splitting
scenedetect[opencv]

# for llava video / captionning
pillow
pillow-avif-plugin
polars
einops
open_clip_torch
av==14.1.0
git+https://github.com/LLaVA-VL/LLaVA-NeXT.git

# for our frontend
gradio==5.20.1
gradio_toggle

# used for the monitor
matplotlib