github/ColossalAI
1
0
Fork 0
mirror of https://github.com/hpcaitech/ColossalAI.git synced 2025-06-28 16:28:10 +00:00
Code Issues Packages Projects Releases Wiki Activity

[hotfix] fix stable diffusion inference bug. (#5289)

Browse Source 
* Update train_ddp.yaml

delete  "strategy" to fix DDP config loading bug in "main.py"

* Update train_ddp.yaml

fix inference with scripts/txt2img.py config file load bug.

* Update README.md

add pretrain model test code.
This commit is contained in:
Youngon 2024-03-05 22:03:40 +08:00 committed by GitHub
parent c8003d463b
commit 68f55a709c
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
2 changed files with 3 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
examples/images/diffusion/README.md
View File

@ -202,6 +202,8 @@ python main.py --logdir /tmp/ -t -b configs/Teyvat/train_colossalai_teyvat.yaml
``` ```
## Inference ## Inference
if you want to test with pretrain model,as bellow:
python scripts/txt2img.py --prompt "a photograph of an astronaut riding a horse" --plms --outdir ./output --ckpt 512-base-ema.ckpt --config configs/train_ddp.yaml
You can get your training last.ckpt and train config.yaml in your `--logdir`, and run by You can get your training last.ckpt and train config.yaml in your `--logdir`, and run by
``` ```

3
examples/images/diffusion/configs/train_ddp.yaml
View File

@ -1,5 +1,6 @@
model: model:
base_learning_rate: 1.0e-4 base_learning_rate: 1.0e-4
target: ldm.models.diffusion.ddpm.LatentDiffusion
params: params:
parameterization: "v" parameterization: "v"
linear_start: 0.00085 linear_start: 0.00085
@ -86,8 +87,6 @@ lightning:
max_epochs: 2 max_epochs: 2
precision: 16 precision: 16
auto_select_gpus: False auto_select_gpus: False
strategy:
find_unused_parameters: False
log_every_n_steps: 2 log_every_n_steps: 2
# max_steps: 6o # max_steps: 6o
logger: True logger: True