Logic/DDT
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

update README

Browse Source
This commit is contained in:
wangshuai6
2025-04-09 11:43:44 +08:00
parent 426803f15b
commit 8e37053625
1 changed files with 9 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

9
README.md
View File

@@ -7,7 +7,16 @@ We decouple diffusion transformer into encoder-decoder design, and surpresingly
* As a byproduct, our DDT can reuse encoder among adjacent steps to accelerate inference. * As a byproduct, our DDT can reuse encoder among adjacent steps to accelerate inference.
## Visualizations ## Visualizations
![](./figs/teaser.png) ![](./figs/teaser.png)
## Checkpoints
Waiting for release.
## Usgae ## Usgae
We use ADM evaluation suite to report FID.
```bash
# for installation
pip install -r requirements.txt
```
```bash ```bash
# for training # for training
python main.py fit -c configs/repa_improved_ddt_xlen22de6_256.yaml python main.py fit -c configs/repa_improved_ddt_xlen22de6_256.yaml