deepgeek/MuseTalk
2
0
Fork 0
mirror of https://github.com/TMElyralab/MuseTalk.git synced 2026-02-05 09:59:18 +08:00
Code Issues Packages Projects Releases Wiki Activity

Update train_codes/README.md

Browse Source
This commit is contained in:
czk32611
2024-04-30 15:06:50 +08:00
parent d73daf1808
commit 30dcd5237f
2 changed files with 11 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

10
train_codes/README.md
View File

@@ -2,6 +2,10 @@
We provde the draft training codes here. Unfortunately, data preprocessing code is still being reorganized.
## Setup
We trained our model on an NVIDIA A100 with `batch size=8, gradient_accumulation_steps=4` for 20w+ steps. Using multiple GPUs should accelerate the training.
## Data preprocessing
You could refer the inference codes which [crop the face images](https://github.com/TMElyralab/MuseTalk/blob/main/scripts/inference.py#L79) and [extract audio features](https://github.com/TMElyralab/MuseTalk/blob/main/scripts/inference.py#L69).
@@ -32,4 +36,8 @@ Finally, the data should be organized as follows:
Simply run after preparing the preprocessed data
```
sh train.sh
```
```
## TODO
- [ ] release data preprocessing codes
- [ ] release some novel designs in training (after technical report)