Fix dtype handling for Adam and LAMB optimizers in 64bit mode. #965
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
This prevents accidental upcasting of parameters due to wonky JAX integer + floating-point py-scalar + array dtype rules.
levskaya
force-pushed
the
optimizer_fixes
branch
from
118fc35 to
dc841b0
Compare
marcvanzee
added
the
Priority: P1 - soon
avital
approved these changes
Feb 2, 2021
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Outdated suggestions cannot be applied.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.
This prevents accidental upcasting of parameters due to wonky JAX integer + floating-point py-scalar + array dtype rules:
in jax x64 mode:
jnp.array(1) + 1.becomes a jax array with dtype float64, that dtype then "invades" the params in the optimizer, causing dtype instability.Should fix #924