This release includes support for quantization of all the Bayesian Convolutional layers listed below in addition to Conv2dReparameterization and Conv2dFlipout.
Conv1dReparameterization,
Conv3dReparameterization,
ConvTranspose1dReparameterization,
ConvTranspose2dReparameterization,
ConvTranspose3dReparameterization,
Conv1dFlipout,
Conv3dFlipout,
ConvTranspose1dFlipout,
ConvTranspose2dFlipout,
ConvTranspose3dFlipout
This release also includes the fixes for the following issues:
Issue #27
Issue #21
Issue #24
Issue #34
What's Changed
- Add quant prepare functions 342ca39
- Fix bug in post-training quantization evaluation due to Jit trace f5c7126
- Add quantization example for ImageNet/ResNet-50 3e74914
- Correcting the order of group and dilation parameters in Conv transpose layers 97ba16a
Full Changelog: v0.4.0...v0.5.0