Feature Request: Add Mish activation

[digantamisra98]

Mish is a new novel activation function proposed in this paper.
It has shown promising results so far and has been adopted in several packages including:

• TensorFlow-Addons
• SpaCy (Tok2Vec Layer)
• Thinc - SpaCy's official NLP based ML library
• Echo AI
• Eclipse's deeplearning4j
• Hasktorch
• CNTKX - Extension of Microsoft's CNTK
• FastAI-Dev
• Darknet
• Yolov3
• BeeDNN - Library in C++
• Gen-EfficientNet-PyTorch
• dnet

All benchmarks, analysis and links to official package implementations can be found in this repository

It would be nice to have Mish as an option within the activation function group.

This is the comparison of Mish with other conventional activation functions in a SEResNet-50 for CIFAR-10: (Better accuracy and faster than GELU)

[hughperkins]

Pull request please :)

…

On Thu, Nov 21, 2019, 09:47 Diganta Misra ***@***.***> wrote: Mish is a new novel activation function proposed in this paper <https://arxiv.org/abs/1908.08681>. It has shown promising results so far and has been adopted in several packages including: - TensorFlow-Addons <https://github.com/tensorflow/addons/tree/master/tensorflow_addons/activations> - SpaCy (Tok2Vec Layer) <https://github.com/explosion/spaCy> - Thinc - SpaCy's official NLP based ML library <https://github.com/explosion/thinc/releases/tag/v7.3.0> - Echo AI <https://github.com/digantamisra98/Echo> - Eclipse's deeplearning4j <deeplearning4j/deeplearning4j#8417> - Hasktorch <https://github.com/hasktorch/hasktorch/blob/e5d4fb1b11663de3e032941f78b9d25c5da06f6d/hasktorch/src/Torch/Typed/Native.hs#L314> - CNTKX - Extension of Microsoft's CNTK <https://github.com/delzac/cntkx> - FastAI-Dev <https://github.com/fastai/fastai_dev/blob/0f613ba3205990c83de9dba0c8798a9eec5452ce/dev/local/layers.py#L441> - Darknet <AlexeyAB/darknet@bf8ea41> - Yolov3 <ultralytics/yolov3@444a9f7> - BeeDNN - Library in C++ <https://github.com/edeforas/BeeDNN> - Gen-EfficientNet-PyTorch <https://github.com/rwightman/gen-efficientnet-pytorch> - dnet <https://github.com/umangjpatel/dnet/blob/master/nn/activations.py> All benchmarks, analysis and links to official package implementations can be found in this repository <https://github.com/digantamisra98/Mish> It would be nice to have Mish as an option within the activation function group. This is the comparison of Mish with other conventional activation functions in a SEResNet-50 for CIFAR-10: (Better accuracy and faster than GELU) [image: se50_1] <https://user-images.githubusercontent.com/34192716/69002745-0de37980-091b-11ea-87da-ac8d17c79e07.png> — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#152?email_source=notifications&email_token=AAA6FKEVHRELFRSPKVV4BWDQU2NOJA5CNFSM4JQDNCR2YY3PNVWWK3TUL52HS4DFUVEXG43VMWVGG33NNVSW45C7NFSM4H3D42MQ>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAA6FKEHZ6IZSBZHYZRURC3QU2NOJANCNFSM4JQDNCRQ> .

[digantamisra98]

@hughperkins will set up one in the next couple of days. Thanks for considering.

[hughperkins]

Thank you :)

…

On Thu, Nov 21, 2019, 09:58 Diganta Misra ***@***.***> wrote: @hughperkins <https://github.com/hughperkins> will set up one in the next couple of days. Thanks for considering. — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#152?email_source=notifications&email_token=AAA6FKAPXH3PKHYYYUBZA5DQU2OYHA5CNFSM4JQDNCR2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEE2QDCA#issuecomment-557121928>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAA6FKDXUHNUJRJD3ESR6WDQU2OYHANCNFSM4JQDNCRQ> .