mirror of
https://github.com/THUDM/CogVideo.git
synced 2025-04-05 19:41:59 +08:00
docs: update finetune documentation in all languages
This commit is contained in:
parent
c817e7f062
commit
ffb6ee36b4
@ -46,7 +46,7 @@ If you need to use a validation dataset during training, make sure to provide a
|
||||
|
||||
Before starting the training, please note the following resolution requirements:
|
||||
|
||||
1. The number of frames must be a multiple of 8.
|
||||
1. The number of frames must be a multiple of 8 **plus 1** (i.e., 8N+1), such as 49, 81, etc.
|
||||
2. The recommended resolution for videos is:
|
||||
- CogVideoX: 480x720 (Height x Width)
|
||||
- CogVideoX1.5: 768x1360 (Height x Width)
|
||||
|
@ -46,7 +46,7 @@ pip install -e .
|
||||
|
||||
トレーニングを開始する前に、以下の解像度設定に関する要件に注意してください:
|
||||
|
||||
1. フレーム数は8の倍数でなければなりません。
|
||||
1. フレーム数は8の倍数 **+1** (つまり8N+1)でなければなりません。例えば49、81など。
|
||||
2. 推奨される動画の解像度は次の通りです:
|
||||
- CogVideoX: 480x720(高さ x 幅)
|
||||
- CogVideoX1.5: 768x1360(高さ x 幅)
|
||||
|
@ -46,7 +46,7 @@ pip install -e .
|
||||
|
||||
在开始训练之前,请注意以下分辨率设置要求:
|
||||
|
||||
1. 帧数必须是8的倍数
|
||||
1. 帧数必须是8的倍数 **+1** (即8N+1), 例如49, 81 ...
|
||||
2. 视频分辨率建议使用模型的默认大小:
|
||||
- CogVideoX: 480x720 (高x宽)
|
||||
- CogVideoX1.5: 768x1360 (高x宽)
|
||||
@ -80,8 +80,6 @@ bash accelerate_train_t2v.sh
|
||||
bash accelerate_train_i2v.sh
|
||||
```
|
||||
|
||||
|
||||
|
||||
## 载入微调的模型
|
||||
|
||||
+ 请关注[cli_demo.py](../inference/cli_demo.py) 以了解如何加载微调的模型。
|
||||
|
Loading…
x
Reference in New Issue
Block a user