mirrors/mmselfsup
1
0
Fork 0
mirror of https://github.com/open-mmlab/mmselfsup.git synced 2025-06-03 14:59:38 +08:00
Code Issues Packages Projects Releases Wiki Activity

[Fix] update lr_updater.py links (#354)

Browse Source 
* update lr_updater.py file path

replace out-of-date links with master branch link

* Update code reference to newest version

1. fix typo err of `CosineAnnealingLrUpdaterHook`
2. update code ref links to the newest version (including `CosineAnnealingLrUpdaterHook`, `CyclicLrUpdaterHook`, `CyclicMomentumUpdaterHook`).
This commit is contained in:
~akA,4}3(V 2022-07-20 10:24:15 +08:00 committed by GitHub
parent 256db8c1b6
commit b9647eb72c
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 3 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
docs/en/tutorials/4_schedule.md
View File

@ -48,9 +48,9 @@ lr_config = dict(
...)
```
Then during training, the program will call [CosineAnealingLrUpdaterHook](https://github.com/open-mmlab/mmcv/blob/f48241a65aebfe07db122e9db320c31b685dc674/mmcv/runner/hooks/lr_updater.py#L227) periodically to update the learning rate.
Then during training, the program will call [CosineAnnealingLrUpdaterHook](https://github.com/open-mmlab/mmcv/blob/0092699fef27a0e6cbe9c37f4c4de2fb6996a1c7/mmcv/runner/hooks/lr_updater.py#L269) periodically to update the learning rate.
We also support many other learning rate schedules [here](https://github.com/open-mmlab/mmcv/blob/f48241a65aebfe07db122e9db320c31b685dc674/mmcv/runner/hooks/lr_updater.py), such as Poly schedule.
We also support many other learning rate schedules [here](https://github.com/open-mmlab/mmcv/blob/master/mmcv/runner/hooks/lr_updater.py), such as Poly schedule.
### Warmup strategy
@ -94,7 +94,7 @@ lr_config = dict(
We support the momentum scheduler to modify the model's momentum according to learning rate, which could make the model converge in a faster way.
Momentum scheduler is usually used with LR scheduler, for example, the following config is used to accelerate convergence. For more details, please refer to the implementation of [CyclicLrUpdater](https://github.com/open-mmlab/mmcv/blob/f48241a65aebfe07db122e9db320c31b685dc674/mmcv/runner/hooks/lr_updater.py#L327) and [CyclicMomentumUpdater](https://github.com/open-mmlab/mmcv/blob/f48241a65aebfe07db122e9db320c31b685dc674/mmcv/runner/hooks/momentum_updater.py#L130).
Momentum scheduler is usually used with LR scheduler, for example, the following config is used to accelerate convergence. For more details, please refer to the implementation of [CyclicLrUpdaterHook](https://github.com/open-mmlab/mmcv/blob/0092699fef27a0e6cbe9c37f4c4de2fb6996a1c7/mmcv/runner/hooks/lr_updater.py#L434) and [CyclicMomentumUpdaterHook](https://github.com/open-mmlab/mmcv/blob/0092699fef27a0e6cbe9c37f4c4de2fb6996a1c7/mmcv/runner/hooks/momentum_updater.py#L291).
Here is an example: