44a19cd37c
Basically: stylegan2 makes use of gradient-based normalizers. These make it so that I cannot use gradient checkpointing. But I love gradient checkpointing. It makes things really, really fast and memory conscious. So - only don't checkpoint when we run the regularizer loss. This is a bit messy, but speeds up training by at least 20%. Also: pytorch: please make checkpointing a first class citizen. |
||
|---|---|---|
| .. | ||
| .idea | ||
| data | ||
| metrics | ||
| models | ||
| scripts | ||
| switched_conv@cb520afd4d | ||
| utils | ||
| multi_modal_train.py | ||
| process_video.py | ||
| requirements.txt | ||
| test.py | ||
| train2.py | ||
| train.py | ||