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.

Sign up for GitHub

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

Jump to bottom

[ci] feat: add more CI workflow #38

Merged
merged 73 commits into from
Jan 9, 2025
Merged

[ci] feat: add more CI workflow #38

merged 73 commits into from
Jan 9, 2025

Conversation

PeterSH6
Copy link
Collaborator

@PeterSH6 PeterSH6 commented Dec 6, 2024

  • As titled

PeterSH6 added 30 commits December 6, 2024 21:51
@PeterSH6
[ci] upload several tests
f4ee653
@PeterSH6
[ci] add sanity and tensordict utility workflow
008a73c
@PeterSH6
[ci] fix workflow
7aa51c0
@PeterSH6
try fix import ci
3c729fd
@PeterSH6
[dataproto] update repeat and unpad/pad
7425b36
@PeterSH6
fix rollout test to 2GPU
23cffb4
@PeterSH6
merge master
7865830
@PeterSH6
add a fsdp vllm hybridengine script, which can be launched by torchrun
3357d99
@PeterSH6
fix import test
8734166
@PeterSH6
merge master
de3b72e
@PeterSH6
update requirement.txt
675bff5
@PeterSH6
draft vllm fsdp test
cc99c80
@PeterSH6
update label
9756354
@PeterSH6
fix
dc205f2
@PeterSH6
upload conda
79ae3b0
@PeterSH6
test conda
2f2bf3d
@PeterSH6
test ci
9bb2cbe
@PeterSH6
use docker
6efecc4
@PeterSH6
test ci
a4bd1cd
@PeterSH6
test ci
1b713fb
@PeterSH6
test ci
5e173a4
@PeterSH6
update ci
5f541cb
@PeterSH6
test ci
3e63cdd
@PeterSH6
fix model loader
e762e98
@PeterSH6
fix model loader
35f3b74
@PeterSH6
test ci
3c2f36e
@PeterSH6
test
9f8e7db
@PeterSH6
upload e2e digit completion test
d47180a
@PeterSH6
update running script for e2e test
902bbde
@PeterSH6
update test config
bfc8adc
vermouth1992
vermouth1992 reviewed Jan 8, 2025
View reviewed changes
verl/protocol.py Outdated Show resolved Hide resolved
@PeterSH6 PeterSH6 changed the title [WIP][ci] add more CI workflow [ci] add more CI workflow Jan 8, 2025
@vermouth1992 vermouth1992 changed the title [ci] add more CI workflow [ci] feat: add more CI workflow Jan 8, 2025
@PeterSH6
Copy link
Collaborator Author

PeterSH6 commented Jan 8, 2025 •
edited
Loading

Nice! Finally...

Some TODOs:

  • Update and fix the colocated worker name in the log and relevant CI
  • Install cupy in L20 using docker.
  • Check the connection error in the ignore ray task in L20.

We can merge this PR first

@eric-haibin-lin
Copy link
Collaborator

so this PR run e2e test with the digit completion trainer. We do not have e2e tests covering the verl ppo trainer yet?

@vermouth1992
Copy link
Collaborator

so this PR run e2e test with the digit completion trainer. We do not have e2e tests covering the verl ppo trainer yet?

It does cover the verl ppo trainer, but not the main entry. It's better to add QWen 0.5b as a nightly ci later

@eric-haibin-lin eric-haibin-lin merged commit c7bd252 into main Jan 9, 2025
6 checks passed
@eric-haibin-lin eric-haibin-lin deleted the gm/ci branch January 9, 2025 01:10
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@vermouth1992 vermouth1992 vermouth1992 left review comments

@eric-haibin-lin eric-haibin-lin eric-haibin-lin approved these changes

Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
3 participants
@PeterSH6 @eric-haibin-lin @vermouth1992