MAML++: BNRS

[DubiousCactus]

Description

Implementation of a BatchNorm layer with Per-Step Batch Normalisation Running Statistics and Per-Step Batch Normalisation Weights and Biases, as proposed in MAML++ by Antobiou et al.

An extension of Pytorch's BatchNorm layer, with the Per-Step Batch Normalisation Running Statistics and Per-Step Batch Normalisation Weights and Biases improvements proposed MAML++ by Antoniou et al. It is adapted from the original Pytorch implementation https://github.com/AntreasAntoniou/with heavy refactoring and a bug (https://github.com/AntreasAntoniou/HowToTrainYourMAMLPytorch/issues/42).

Contribution Checklist

If your contribution modifies code in the core library (not docs, tests, or examples), please fill the following checklist.

• [x] My contribution is listed in CHANGELOG.md with attribution.
• [x] My contribution modifies code in the main library.
• [ ] My modifications are tested.
• [x] My modifications are documented.

[lgtm-com[bot]]

This pull request introduces 3 alerts when merging 77eff55153cfda4bbbbe0ec74bcb4b4fe23f225e into df3c3291b4681440a80a69a7815090a4bd3cd661 - view on LGTM.com

new alerts:

• 3 for Unused import

[lgtm-com[bot]]

This pull request introduces 2 alerts when merging 2beb0b2bee3e57695f77162a1fae9efacf3fdd01 into df3c3291b4681440a80a69a7815090a4bd3cd661 - view on LGTM.com

new alerts:

• 1 for Unused local variable
• 1 for Variable defined multiple times