use instructlab constraints-dev.txt in e2e test
#499
Merged
Conversation
This file contains hidden or 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
For instructlab, "pip install ." does not install vllm, but it does install an uncapped torch (2.7.0 currently). When we install vllm later, we compile a binary flash_attn wheel against torch 2.7.0. vllm 0.8.4 requires torch==2.6.0, so we downgrade torch, and then we use that with the incompatible flash_attn binary wheel. To resolve this, use instructlab's constraints-dev.txt in the first pip install operation. This restricts torch to 2.6.0 immediately when we first install instructlab, so that we will compile flash_attn against that torch version. Signed-off-by: Ken Dreyer <[email protected]>
ktdreyer
force-pushed
the
e2e-use-constraints
branch
from
e31adef to
b771fc7
Compare
|
E2E (NVIDIA L40S x4) workflow launched on this PR: View run |
|
e2e workflow failed on this PR: View run, please investigate. |
Contributor
Author
|
This workflow file was copied to this training repository from https://github.com/instructlab/instructlab/blob/main/.github/workflows/e2e-nvidia-l40s-x4.yml, so this problem exists there also. I'm fixing that in instructlab/instructlab#3320 |
booxter
approved these changes
Apr 28, 2025
Contributor
booxter
left a comment
booxter
left a comment
There was a problem hiding this comment.
Correct fix. Smoke tests are affected by something else (even if adjacent.)
JamesKunstle
approved these changes
Apr 28, 2025
Contributor
Author
|
The E2E (NVIDIA L40S x4) test still fails about three hours in, in NCCL timeout errors. I've filed instructlab/instructlab#3321 to track that problem. |
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Outdated suggestions cannot be applied.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.
For instructlab,
pip install .does not installvllm, but it does install an uncappedtorch(2.7.0currently).When we install
vllmlater, we compile a binaryflash_attnwheel againsttorch-2.7.0.vllm-0.8.4requirestorch==2.6.0, so we downgradetorch, and then we use that with the incompatibleflash_attnbinary wheel.To resolve this, use instructlab's
constraints-dev.txtin the firstpip installoperation. This restrictstorchto2.6.0immediately when we first install instructlab, so that we will compileflash_attnagainst thattorchversion.Fixes: #494