Ma Zerun b017670e1b

[Improve] Use PyTorch official scaled_dot_product_attention to accelerate MultiheadAttention. (#1434) ...

* [Improve] Use PyTorch official `scaled_dot_product_attention` to accelerate `MultiheadAttention`.

* Support `--local-rank` and `--amp` option for new version PyTorch.

* Fix imports and UT.

2023-03-29 15:50:44 +08:00

..

batch_augments

Rename the package name to mmpretrain.

2023-02-17 15:20:55 +08:00

__init__.py

[Docs] Update user guides docs and tools for MMPretrain. (#1429)

2023-03-27 14:32:26 +08:00

attention.py

[Improve] Use PyTorch official scaled_dot_product_attention to accelerate MultiheadAttention. (#1434)

2023-03-29 15:50:44 +08:00

batch_shuffle.py

[Refactor] Move and refactor utils from mmselfsup. (#1385)

2023-02-28 17:04:40 +08:00

channel_shuffle.py

Rename the package name to mmpretrain.

2023-02-17 15:20:55 +08:00

clip_generator_helper.py

[Refactor] Add self-supervised backbones and target generators. (#1379)

2023-02-28 15:59:17 +08:00

data_preprocessor.py

[Refactor] Add selfsup algorithms. (#1389)

2023-03-06 16:53:15 +08:00

ema.py

[Refactor] Add selfsup algorithms. (#1389)

2023-03-06 16:53:15 +08:00

embed.py

Rename the package name to mmpretrain.

2023-02-17 15:20:55 +08:00

helpers.py

Rename the package name to mmpretrain.

2023-02-17 15:20:55 +08:00

inverted_residual.py

Rename the package name to mmpretrain.

2023-02-17 15:20:55 +08:00

layer_scale.py

Rename the package name to mmpretrain.

2023-02-17 15:20:55 +08:00

make_divisible.py

Rename the package name to mmpretrain.

2023-02-17 15:20:55 +08:00

norm.py

[Docs] Update user guides docs and tools for MMPretrain. (#1429)

2023-03-27 14:32:26 +08:00

position_encoding.py

[Refactor] Use out_type to specify ViT-like backbone output. (#1408)

2023-03-09 11:02:58 +08:00

res_layer_extra_norm.py

[Docs] Update user guides docs and tools for MMPretrain. (#1429)

2023-03-27 14:32:26 +08:00

se_layer.py

Rename the package name to mmpretrain.

2023-02-17 15:20:55 +08:00

vector_quantizer.py

[Refactor] Add self-supervised backbones and target generators. (#1379)

2023-02-28 15:59:17 +08:00