awslabs / awslabs/slapo

[Test] Unit test robustness

Open
#11 2 comments 0 reactions 1 assignee Claimed by @chhzh123 View on GitHub
Dominant language
Python
Stars
153
Forks
17
PR merge metrics
No merged PRs in 30d

Description

Since the CI is set up., we now use this issue to track the progress of adding unit tests to make the repo more robustness. Specifically, we will cover the following checklist:

- [x] Temporary end-to-end tests. Add a few end-to-end tests from `examples/` to make sure the changes won't break existing benchmarks. (#14)
- [x] Unit tests - tracer. (#16)
- [x] Unit tests - initialization and consolidation. (#54)
- [ ] Unit tests - schedule primitives
- [x] schedule hierarchy and sub-schedule accessing. (#29)
- [x] tracing. (#29)
- [x] sharding and syncing.
- [x] find node and subgraph pattern matching. (#35)
- [x] replace function and module. (#52)
- [x] operator fusion. (#25)
- [x] activation checkpointing (module and subgraph). (#55)
- [ ] pipeline partitioning.
- [ ] Unit tests - auto-tuning.
- [ ] Unit tests - DeepSpeed dialect.
- [ ] Remove temporary end-to-end tests.

Anyone could claim the above tasks by commenting this issue with an ETA. Please also link the submitted PRs to this issue for tracking.

Contributor guide

Open the contributing guide

Assessment

This issue has not been assessed yet.

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.