Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Fix converting only float type tensors in Lite #10429

Merged
merged 9 commits into from
Nov 9, 2021

Conversation

awaelchli
Copy link
Contributor

@awaelchli awaelchli commented Nov 9, 2021

What does this PR do?

Fixes #10371

Lite converts inputs automatically to the target precision (32, 16, bf16). However, so far this happened unconditionally for all tensor types, when in reality this only makes sense for floating point tensors. This PR restricts this conversion to float tensors only, both for input and output of the wrapped model-forward.

Test fails on master (i.e., without the fix applied).

Does your PR introduce any breaking changes? If yes, please list them.

No

Before submitting

  • Was this discussed/approved via a GitHub issue? (not for typos and docs)
  • Did you read the contributor guideline, Pull Request section?
  • Did you make sure your PR does only one thing, instead of bundling different changes together?
  • Did you make sure to update the documentation with your changes? (if necessary)
  • Did you write any new necessary tests? (not for typos and docs)
  • Did you verify new and existing tests pass locally with your changes?
  • Did you update the CHANGELOG? (not for typos, docs, test updates, or internal minor changes/refactorings)

PR review

Anyone in the community is free to review the PR once the tests have passed.
Before you start reviewing make sure you have read Review guidelines. In short, see the following bullet-list:

  • Is this pull request ready for review? (if not, please submit in draft mode)
  • Check that all items from Before submitting are resolved
  • Make sure the title is self-explanatory and the description concisely explains the PR
  • Add labels and milestones (and optionally projects) to the PR so it can be classified

Did you have fun?

I made sure I had fun coding 🙃

Part of #1 (it's a lie, this is just here to avoid noisy GitHub bot)

@awaelchli awaelchli added bug Something isn't working fabric lightning.fabric.Fabric labels Nov 9, 2021
@awaelchli awaelchli added this to the 1.5.x milestone Nov 9, 2021
@awaelchli awaelchli added the priority: 0 High priority task label Nov 9, 2021
@awaelchli awaelchli marked this pull request as ready for review November 9, 2021 10:37
@mergify mergify bot added the ready PRs ready to be merged label Nov 9, 2021
@awaelchli awaelchli mentioned this pull request Nov 9, 2021
11 tasks
@mergify mergify bot added the has conflicts label Nov 9, 2021
@mergify mergify bot removed the has conflicts label Nov 9, 2021
@justusschock justusschock enabled auto-merge (squash) November 9, 2021 12:57
Copy link
Contributor

@carmocca carmocca left a comment

Choose a reason for hiding this comment

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

We'll need to iterate on this but it's okay for now

pytorch_lightning/lite/wrappers.py Show resolved Hide resolved
@justusschock justusschock merged commit aaa6aa7 into master Nov 9, 2021
@justusschock justusschock deleted the bugfix/lite-precision-conversion branch November 9, 2021 14:21
awaelchli added a commit that referenced this pull request Nov 9, 2021
* fix

* less code

* add test case

* add test cases

* update input

* add test cases

* add type hint

* add changelog note

Co-authored-by: Kaushik B <[email protected]>
awaelchli added a commit that referenced this pull request Nov 9, 2021
* fix

* less code

* add test case

* add test cases

* update input

* add test cases

* add type hint

* add changelog note

Co-authored-by: Kaushik B <[email protected]>
lexierule pushed a commit that referenced this pull request Nov 9, 2021
* fix

* less code

* add test case

* add test cases

* update input

* add test cases

* add type hint

* add changelog note

Co-authored-by: Kaushik B <[email protected]>
Raalsky pushed a commit to neptune-ai/pytorch-lightning that referenced this pull request Nov 23, 2021
* fix

* less code

* add test case

* add test cases

* update input

* add test cases

* add type hint

* add changelog note

Co-authored-by: Kaushik B <[email protected]>
@rohitgr7 rohitgr7 mentioned this pull request Feb 7, 2022
12 tasks
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
bug Something isn't working fabric lightning.fabric.Fabric priority: 0 High priority task ready PRs ready to be merged
Projects
None yet
Development

Successfully merging this pull request may close these issues.

Why does bf16 of LightningLite use GPU Memory much more than pytorch_lightning.Trainer
6 participants