Add support for Bayesian neural network quantization #23
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Signed-off-by: Ranganath Krishnan <ranganath.krishnan@intel.com>
Signed-off-by: Ranganath Krishnan <ranganath.krishnan@intel.com>
Signed-off-by: Ranganath Krishnan <ranganath.krishnan@intel.com>
Signed-off-by: Ranganath Krishnan <ranganath.krishnan@intel.com>
to compute kl when 'return_kl' flag is set to False. Fix for issue#12. Signed-off-by: Ranganath Krishnan <ranganath.krishnan@intel.com>
update usage instructions in README file
Signed-off-by: Ranganath Krishnan <ranganath.krishnan@intel.com>
* fix minor typo. Signed-off-by: Ranganath Krishnan <ranganath.krishnan@intel.com> * Update links in README.md * update MOPED layer example utility function Signed-off-by: Ranganath Krishnan <ranganath.krishnan@intel.com> * Update README.md * feat: add possibility to return no kl, save it as attribute * feat: add possibility to return no kl on flipout layers, save it as attribute * updates to support dnn to bnn imodel auto conversion * updates to support dnn to bnn imodel auto conversion * remove duplicate kl_loss definition in Conv1dReparameterization layer Signed-off-by: Ranganath Krishnan <ranganath.krishnan@intel.com> * include kl_loss() function in Convolutional flipout layers, to compute kl when 'return_kl' flag is set to False. Fix for issue#12. Signed-off-by: Ranganath Krishnan <ranganath.krishnan@intel.com> * Update README.md * Update README.md * update the posterior variational param init value Signed-off-by: Ranganath Krishnan <ranganath.krishnan@intel.com> * Update release version with dnn_to_bnn() feature * Update README.md update usage instructions in README file * Update requirements.txt * Include training, testing and uncertainty quantification snippet in README.md * update version in setup.py * Update bayesian_torch.layers.md * Update links in README.md * Update setup.py * Update README.md * include assets folder Signed-off-by: Ranganath Krishnan <ranganath.krishnan@intel.com> * Update README.md * Update README.md * Update README.md * Update README.md * Update setup.py * release to PyPI, update install instruction through "pip" command * Switched to permanent URL for the top image. * changing to raw.githubusercontent.com Url for top image. * Update README.md * update links and release number for PyPI documentation Signed-off-by: Ranganath Krishnan <ranganath.krishnan@intel.com> * Update README.md add downloads statistics badge * update download count badge * Added support for arbitrary kernel sizes for Bayesian Conv layers * update version number Signed-off-by: Ranganath Krishnan <ranganath.krishnan@intel.com> * Update README.md * Add support for output padding in flipout layers --------- Signed-off-by: Ranganath Krishnan <ranganath.krishnan@intel.com> Co-authored-by: Ranganath Krishnan <ranganath.krishnan@intel.com> Co-authored-by: Pi <piero.skywalker@gmail.com> Co-authored-by: msubedar <mahesh.subedar@intel.com> Co-authored-by: Michael Beale <michael.beale@intel.com>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
No description provided.