limuloo / PyDIff

[IJCAI 2023 ORAL] "Pyramid Diffusion Models For Low-light Image Enhancement" (Official Implementation)
Other
149 stars 8 forks source link

time #4

Closed zhenyuan1234 closed 1 year ago

zhenyuan1234 commented 1 year ago

How long did it take to train the model?

limuloo commented 1 year ago

According to the configuration in the paper, it takes 4 days to train on 2 3090s, but I have a faster configuration for experiments, and it takes about 1 day. If necessary, you can contact my email address zdw1999@zju.edu.cn to apply for a complete training code.

finnn00 commented 3 weeks ago

Hello Can I get the faster configuration for experiments? Now I am working on the condition of gpu=1, batch size=4, May I ask how many days it would take to retrain your model? Also can I get the FPS code too? Looking for ur reply! My email: finn951017@outlook.com

limuloo commented 3 weeks ago

@finnn00 To enable PyDiff to train on a single GPU, the PyDiff/options/train_v2.yaml file can be consulted.