[Auto]Fix loading gpu trained estimator on cpu machine #1717
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
yinweisu
changed the title
|
Job PR-1717-994dda3 is done. |
|
Job PR-1717-ef708ec is done. |
Merged
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Outdated suggestions cannot be applied.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.
Issue #:
autogluon/autogluon#1426
Description:
autogluon users reported a bug where predictor trained on GPU machine is not loadable on a CPU machine.
The previous logic of loading uses pickle, which loads object from bottom down and internally calls torch.load before reaching TorchImageClassificationEstimator custom loading logic.
After refactor loading logic to use torch.load, this issue is solved.
Ideally, unit test should be added to cover such scenario. However, there's no way to disable gpu after torch has been loaded. Hence, impossible to test it on a single machine.