mmpretrain.engine¶
This package includes some runtime components, including hooks, runners, optimizers and loops. These components are useful in classification tasks but not supported by MMEngine yet.
Note
Some components may be moved to MMEngine in the future.
Hooks¶
Class Number Check HOOK. |
|
Precise BN hook. |
|
Classification Visualization Hook. |
|
The hook to prepare the prototype in retrievers. |
|
Set adaptive-margins in ArcFaceClsHead based on the power of category-wise count. |
|
A Hook to apply Exponential Moving Average (EMA) on the model during training. |
|
Hook for SimSiam. |
|
Hook for DenseCL. |
|
Hook for SwAV. |
Optimizers¶
A pure pytorch variant of FuseLAMB (NvLamb variant) optimizer. |
|
Implements layer-wise adaptive rate scaling for SGD. |
|
Different learning rates are set for different layers of backbone. |