Download configs/backup/common/lsj-200e_coco-instance.py from scfive/samewind: direct link, hf CLI and curl.
- Browser
- Download file 379 Bytes
-
https://huggingface.co/spaces/scfive/samewind/resolve/main/configs/backup/common/lsj-200e_coco-instance.py
- Command line
-
hf download hf://spaces/scfive/samewind/configs/backup/common/lsj-200e_coco-instance.py
-
curl -L -o lsj-200e_coco-instance.py https://huggingface.co/spaces/scfive/samewind/resolve/main/configs/backup/common/lsj-200e_coco-instance.py
379 Bytes
| _base_ = './lsj-100e_coco-instance.py' | |
| # 8x25=200e | |
| train_dataloader = dict(dataset=dict(times=8)) | |
| # learning rate | |
| param_scheduler = [ | |
| dict( | |
| type='LinearLR', start_factor=0.067, by_epoch=False, begin=0, | |
| end=1000), | |
| dict( | |
| type='MultiStepLR', | |
| begin=0, | |
| end=25, | |
| by_epoch=True, | |
| milestones=[22, 24], | |
| gamma=0.1) | |
| ] | |