Skip to content

[3/N][CI/UT] add spec decode e2e UT #487

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.

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

Open
wants to merge 4 commits into
base: v0.7.3-dev
Choose a base branch
from

Conversation

mengwei805
Copy link

@mengwei805 mengwei805 commented Apr 9, 2025

What this PR does / why we need it?

  1. add test_eagle_correctness.py; use local model weights;
  2. add test_mtp_correctness.py; use bf16 model weights;

Does this PR introduce any user-facing change?

None

How was this patch tested?

Local verification passed

@mengwei805 mengwei805 force-pushed the v0.7.3-sd-ut-part3 branch 7 times, most recently from 63f049c to 7c03c68 Compare April 11, 2025 06:09
Co-authored-by: mengwei805 <mengwei25@huawei.com>
Co-authored-by: MengqingCao <cmq0113@163.com>
Signed-off-by: mengwei805 <mengwei25@huawei.com>
@@ -148,7 +148,7 @@ jobs:
- name: Run vllm-project/vllm-ascend key feature test
if: steps.filter.outputs.speculative_tests_changed
run: |
pytest -sv tests/spec_decode
pytest -sv tests/spec_decode/e2e/test_eagle_correctness.py::test_llama2_eagle_e2e_greedy_correctness
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

todo: Revert this when successfully download llama model

Copy link
Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

OK, the modification here is just for debugging. I will restore it after debugging is successful.

Signed-off-by: mengwei805 <mengwei25@huawei.com>
Signed-off-by: mengwei805 <mengwei25@huawei.com>
Signed-off-by: mengwei805 <mengwei25@huawei.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants