Support for SeparableConv1DFlipout

tensorflow / probability

Probabilistic reasoning and statistical analysis in TensorFlow

https://www.tensorflow.org/probability/

Apache License 2.0

4.23k stars 1.09k forks source link

Support for SeparableConv1DFlipout #745

Open jackshi0912 opened 4 years ago

jackshi0912 commented 4 years ago

SeparableConv enables a much greater range of low power, low latency models. MobileNet and other low cost high performance networks generally utilizes SeparableConv for it's efficiency. I am building a real probabilistic inference version of a MobileNet which uses SeparableConv. I am using regular tfp convolution ops for now for initial validation but it is much less efficient and will not be fit for production use. Thanks!

abhichou4 commented 4 years ago

Hey @jackshi0912 @srvasude, I'm a beginner on this project and would love to help with this issue. Any tips on how to get started would be much appreciated.

Sarimuko commented 4 years ago

Hi @jackshi0912 @srvasude, I can take this issue, and I will begin to check the code. If there is any suggestion, please let me know.