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This repository has been archived by the owner on Nov 17, 2023.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

Move ops which don't support FP16 dtype to FP32 list #16668

Merged
merged 1 commit into from
Oct 29, 2019

Conversation

anirudh2290
Copy link
Member

Description

Some legacy ops call this InferType function here: https://github.com/apache/incubator-mxnet/blob/master/include/mxnet/operator.h#L228 ,which doesnt support FP16 dtype. Tweaking the AMP lists so that users don't run into issues like the one mentioned here: https://discuss.mxnet.io/t/convert-model-to-float16/5053/3

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

@anirudh2290
Copy link
Member Author

@ptrendx can you please take a look.

@anirudh2290 anirudh2290 changed the title Move ops which don't support FP16 dtype to FP32 list [WIP] Move ops which don't support FP16 dtype to FP32 list Oct 29, 2019
@anirudh2290 anirudh2290 changed the title [WIP] Move ops which don't support FP16 dtype to FP32 list Move ops which don't support FP16 dtype to FP32 list Oct 29, 2019
Copy link
Member

@ptrendx ptrendx left a comment

Choose a reason for hiding this comment

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

LGTM. Why did you remove the BN v1 and Polling v1? Do they support fp16?

@anirudh2290
Copy link
Member Author

anirudh2290 commented Oct 29, 2019

LGTM. Why did you remove the BN v1 and Polling v1? Do they support fp16?

it was a mistake. removed it now. yes they do support FP16, they have their own InferType functions.

@anirudh2290 anirudh2290 merged commit 5aa74e0 into apache:master Oct 29, 2019
yajiedesign pushed a commit to yajiedesign/mxnet that referenced this pull request Nov 6, 2019
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants