open-mmlab/mmsegmentation
Vedi su GitHubToo much GPU memory consumption and seems wrong
Open
#3167 aperta il 6 lug 2023
good first issue
Metriche repository
- Star
- (6867 star)
- Metriche merge PR
- (Nessuna PR mergiata in 30 g)
Descrizione
When I use resnet101 with deeplabv3+ in this lib 'https://github.com/VainF/DeepLabV3Plus-Pytorch', I can train for 40 batch size on 2 GPUs, 20 batch size for each GPU, and it consumes about 36 GB. However, when I use your lib, for 2 batch size on 1 GPUs, it consumes about 15 GB