Skip to content
This repository has been archived by the owner on Nov 17, 2023. It is now read-only.

[Numpy] Random.normal() with backward #16330

Merged
merged 5 commits into from
Nov 18, 2019
Merged

Conversation

xidulu
Copy link
Contributor

@xidulu xidulu commented Sep 29, 2019

Description

As title
Gradient estimated through reparameterization trick:

z ~ N(\mu, \sigma) = N(0, 1) * \sigma + \mu
dz/d\mu = 1
dz/d\sigma = N(0, 1)

Checklist

Essentials

Please feel free to remove inapplicable items for your PR.

  • The PR title starts with [MXNET-$JIRA_ID], where $JIRA_ID refers to the relevant JIRA issue created (except PRs with tiny changes)
  • Changes are complete (i.e. I finished coding on this PR)
  • All changes have test coverage:
  • Unit tests are added for small changes to verify correctness (e.g. adding a new operator)
  • Nightly tests are added for complicated/long-running ones (e.g. changing distributed kvstore)
  • Build tests will be added for build configuration changes (e.g. adding a new build option with NCCL)
  • Code is well-documented:
  • For user-facing API changes, API doc string has been updated.
  • For new C++ functions in header files, their functionalities and arguments are documented.
  • For new examples, README.md is added to explain the what the example does, the source of the dataset, expected performance on test set and reference to the original paper if applicable
  • Check the API doc at https://mxnet-ci-doc.s3-accelerate.dualstack.amazonaws.com/PR-$PR_ID/$BUILD_ID/index.html
  • To the my best knowledge, examples are either not affected by this change, or have been fixed to be compatible with this change

Changes

  • Feature1, tests, (and when applicable, API doc)
  • Feature2, tests, (and when applicable, API doc)

Comments

  • If this change is a backward incompatible change, why must this change be made.
  • Interesting edge cases to note here

@sxjscience
Copy link
Member

You may mention that this uses the reparameterization trick.

@haojin2
Copy link
Contributor

haojin2 commented Nov 17, 2019

@sxjscience This PR looks good to you?

@sxjscience
Copy link
Member

LGTM. We can merge this in once the conflicts are resolved.

@xidulu
Copy link
Contributor Author

xidulu commented Nov 17, 2019

@sxjscience
Sure, I will resolve the conflict ASAP.

Copy link
Contributor

@haojin2 haojin2 left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM

@haojin2 haojin2 merged commit 8cc4c16 into apache:master Nov 18, 2019
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants