ml-testing-accelerators
ml-testing-accelerators copied to clipboard
GoogleCloudPlatform
Update hf-diffusers.libsonnet checkpoint steps to skip save checkpoints
Description
Update hf-diffusers.libsonnet checkpoint steps to skip save checkpoints
Tests
Please describe the tests that you ran on TPUs to verify changes.
Instruction and/or command lines to reproduce your tests: ./scripts/run-oneshot.sh -t pt-2.1-hf-diffusers-conv-v4-8-1vm
List links for your tests (use go/shortn-gen for any internal link): ./scripts/run-oneshot.sh -t pt-2.1-hf-diffusers-conv-v4-8-1vm: http://shortn/_XUJnu7W1FP
Checklist
Before submitting this PR, please make sure (put X in square brackets):
- [x] I have performed a self-review of my code.
- [x] I have necessary comments in my code, particularly in hard-to-understand areas.
- [x] I have run one-shot tests and provided workload links above if applicable.
- [x] I have made or will make corresponding changes to the doc if needed.
Just to get back on track, what was the reason for this change?
The reason of this change is to make the test green, but we are trying to debug the reason of saving checkpoint hang locally too now
Just to get back on track, what was the reason for this change?
The reason of this change is to make the test green, but we are trying to debug the reason of saving checkpoint hang locally too now
Right, to rephrase my question, it was asking the reason why we need to update the checkpoint steps to make the test green. But seems like the investigation is TBD, thanks!