[Bugfix]Fix accuracy test#953
Closed
zhangxinyuehfad wants to merge 1 commit intovllm-project:mainfrom
Closed
Conversation
zhangxinyuehfad
force-pushed
the
zxy_accuracy_workflow
branch
2 times, most recently
from
0636582 to
b4f8d25
Compare
Yikun
reviewed
May 26, 2025
| @@ -28,6 +28,10 @@ on: | |||
| description: 'what vllm-ascend version to accuracy test?' | |||
| required: true | |||
| type: string | |||
| @@ -28,6 +28,10 @@ on: | |||
| description: 'what vllm-ascend version to accuracy test?' | |||
Member
There was a problem hiding this comment.
vLLM version:
vLLM Ascend version:
Model:
zhangxinyuehfad
force-pushed
the
zxy_accuracy_workflow
branch
from
b4f8d25 to
8503890
Compare
wangxiyuan
reviewed
May 28, 2025
| required: true | ||
| type: string | ||
| type: choice | ||
| options: |
Collaborator
There was a problem hiding this comment.
we don't support all vllm version. Please just list the version vllm-ascend support:
| type: string | ||
| type: choice | ||
| options: | ||
| - main |
Collaborator
There was a problem hiding this comment.
0.7.1 is out-of-date, no need to test it.
zhangxinyuehfad
force-pushed
the
zxy_accuracy_workflow
branch
from
8503890 to
53bf706
Compare
Signed-off-by: hfadzxy <starmoon_zhang@163.com>
zhangxinyuehfad
force-pushed
the
zxy_accuracy_workflow
branch
from
53bf706 to
6aa72a5
Compare
Yikun
approved these changes
May 30, 2025
Member
|
Could we add a PR triggered event to make sure the PR work? |
Member
|
Replaced by #1040 |
Yikun
added a commit
that referenced
this pull request
Jun 3, 2025
### What this PR does / why we need it? This PR enable accuracy test for PR labeled with "*accuracy-test" and workflow_dispatch. Only one model test running for each type test to reduce excution time. - The dense test costs about `25mins` to complete (gsm8k 7mins, ~mmlu 3h24mins,~ cEval 18mins) - The vl test costs about `40mins` to complete In futute, we might consider enable all job test as nightly schedule job. Below is mainly changes: - the dense/vl accuracy test will be triggered by lableling `accuracy-test` and `ready-for-test` - the dense accuracy test will be triggered by lableling `dense-accuracy-test` and `ready-for-test` - the vl accuracy test will be triggered by lableling `vl-accuracy-test` and `ready-for-test` - accuracy test will also be triggered by workflow_dispatch - Support V1 and V0 for qwen and V0 for VL For PR test we also generate summary in test summary. ### Does this PR introduce _any_ user-facing change? No ### How was this patch tested? - CI passed with accuracy-test label - Preview: https://github.com/vllm-project/vllm-ascend/actions/runs/15407628722?pr=1040 Closes: #953 --------- Signed-off-by: hfadzxy <starmoon_zhang@163.com> Signed-off-by: Yikun Jiang <yikunkero@gmail.com> Co-authored-by: hfadzxy <starmoon_zhang@163.com>
momo609
pushed a commit
to momo609/vllm-ascend
that referenced
this pull request
Jun 3, 2025
…ct#1040) ### What this PR does / why we need it? This PR enable accuracy test for PR labeled with "*accuracy-test" and workflow_dispatch. Only one model test running for each type test to reduce excution time. - The dense test costs about `25mins` to complete (gsm8k 7mins, ~mmlu 3h24mins,~ cEval 18mins) - The vl test costs about `40mins` to complete In futute, we might consider enable all job test as nightly schedule job. Below is mainly changes: - the dense/vl accuracy test will be triggered by lableling `accuracy-test` and `ready-for-test` - the dense accuracy test will be triggered by lableling `dense-accuracy-test` and `ready-for-test` - the vl accuracy test will be triggered by lableling `vl-accuracy-test` and `ready-for-test` - accuracy test will also be triggered by workflow_dispatch - Support V1 and V0 for qwen and V0 for VL For PR test we also generate summary in test summary. ### Does this PR introduce _any_ user-facing change? No ### How was this patch tested? - CI passed with accuracy-test label - Preview: https://github.com/vllm-project/vllm-ascend/actions/runs/15407628722?pr=1040 Closes: vllm-project#953 --------- Signed-off-by: hfadzxy <starmoon_zhang@163.com> Signed-off-by: Yikun Jiang <yikunkero@gmail.com> Co-authored-by: hfadzxy <starmoon_zhang@163.com> Signed-off-by: wangxiaoxin (A) <w00664509@china.huawei.com>
momo609
pushed a commit
to momo609/vllm-ascend
that referenced
this pull request
Jun 3, 2025
…ct#1040) ### What this PR does / why we need it? This PR enable accuracy test for PR labeled with "*accuracy-test" and workflow_dispatch. Only one model test running for each type test to reduce excution time. - The dense test costs about `25mins` to complete (gsm8k 7mins, ~mmlu 3h24mins,~ cEval 18mins) - The vl test costs about `40mins` to complete In futute, we might consider enable all job test as nightly schedule job. Below is mainly changes: - the dense/vl accuracy test will be triggered by lableling `accuracy-test` and `ready-for-test` - the dense accuracy test will be triggered by lableling `dense-accuracy-test` and `ready-for-test` - the vl accuracy test will be triggered by lableling `vl-accuracy-test` and `ready-for-test` - accuracy test will also be triggered by workflow_dispatch - Support V1 and V0 for qwen and V0 for VL For PR test we also generate summary in test summary. ### Does this PR introduce _any_ user-facing change? No ### How was this patch tested? - CI passed with accuracy-test label - Preview: https://github.com/vllm-project/vllm-ascend/actions/runs/15407628722?pr=1040 Closes: vllm-project#953 --------- Signed-off-by: hfadzxy <starmoon_zhang@163.com> Signed-off-by: Yikun Jiang <yikunkero@gmail.com> Co-authored-by: hfadzxy <starmoon_zhang@163.com> Signed-off-by: wangxiaoxin (A) <w00664509@china.huawei.com>
momo609
pushed a commit
to momo609/vllm-ascend
that referenced
this pull request
Jun 3, 2025
…ct#1040) ### What this PR does / why we need it? This PR enable accuracy test for PR labeled with "*accuracy-test" and workflow_dispatch. Only one model test running for each type test to reduce excution time. - The dense test costs about `25mins` to complete (gsm8k 7mins, ~mmlu 3h24mins,~ cEval 18mins) - The vl test costs about `40mins` to complete In futute, we might consider enable all job test as nightly schedule job. Below is mainly changes: - the dense/vl accuracy test will be triggered by lableling `accuracy-test` and `ready-for-test` - the dense accuracy test will be triggered by lableling `dense-accuracy-test` and `ready-for-test` - the vl accuracy test will be triggered by lableling `vl-accuracy-test` and `ready-for-test` - accuracy test will also be triggered by workflow_dispatch - Support V1 and V0 for qwen and V0 for VL For PR test we also generate summary in test summary. ### Does this PR introduce _any_ user-facing change? No ### How was this patch tested? - CI passed with accuracy-test label - Preview: https://github.com/vllm-project/vllm-ascend/actions/runs/15407628722?pr=1040 Closes: vllm-project#953 --------- Signed-off-by: hfadzxy <starmoon_zhang@163.com> Signed-off-by: Yikun Jiang <yikunkero@gmail.com> Co-authored-by: hfadzxy <starmoon_zhang@163.com> Signed-off-by: wangxiaoxin (A) <w00664509@china.huawei.com>
momo609
pushed a commit
to momo609/vllm-ascend
that referenced
this pull request
Jun 3, 2025
…ct#1040) ### What this PR does / why we need it? This PR enable accuracy test for PR labeled with "*accuracy-test" and workflow_dispatch. Only one model test running for each type test to reduce excution time. - The dense test costs about `25mins` to complete (gsm8k 7mins, ~mmlu 3h24mins,~ cEval 18mins) - The vl test costs about `40mins` to complete In futute, we might consider enable all job test as nightly schedule job. Below is mainly changes: - the dense/vl accuracy test will be triggered by lableling `accuracy-test` and `ready-for-test` - the dense accuracy test will be triggered by lableling `dense-accuracy-test` and `ready-for-test` - the vl accuracy test will be triggered by lableling `vl-accuracy-test` and `ready-for-test` - accuracy test will also be triggered by workflow_dispatch - Support V1 and V0 for qwen and V0 for VL For PR test we also generate summary in test summary. ### Does this PR introduce _any_ user-facing change? No ### How was this patch tested? - CI passed with accuracy-test label - Preview: https://github.com/vllm-project/vllm-ascend/actions/runs/15407628722?pr=1040 Closes: vllm-project#953 --------- Signed-off-by: hfadzxy <starmoon_zhang@163.com> Signed-off-by: Yikun Jiang <yikunkero@gmail.com> Co-authored-by: hfadzxy <starmoon_zhang@163.com> Signed-off-by: wangxiaoxin (A) <w00664509@china.huawei.com>
momo609
pushed a commit
to momo609/vllm-ascend
that referenced
this pull request
Jun 4, 2025
…ct#1040) ### What this PR does / why we need it? This PR enable accuracy test for PR labeled with "*accuracy-test" and workflow_dispatch. Only one model test running for each type test to reduce excution time. - The dense test costs about `25mins` to complete (gsm8k 7mins, ~mmlu 3h24mins,~ cEval 18mins) - The vl test costs about `40mins` to complete In futute, we might consider enable all job test as nightly schedule job. Below is mainly changes: - the dense/vl accuracy test will be triggered by lableling `accuracy-test` and `ready-for-test` - the dense accuracy test will be triggered by lableling `dense-accuracy-test` and `ready-for-test` - the vl accuracy test will be triggered by lableling `vl-accuracy-test` and `ready-for-test` - accuracy test will also be triggered by workflow_dispatch - Support V1 and V0 for qwen and V0 for VL For PR test we also generate summary in test summary. ### Does this PR introduce _any_ user-facing change? No ### How was this patch tested? - CI passed with accuracy-test label - Preview: https://github.com/vllm-project/vllm-ascend/actions/runs/15407628722?pr=1040 Closes: vllm-project#953 --------- Signed-off-by: hfadzxy <starmoon_zhang@163.com> Signed-off-by: Yikun Jiang <yikunkero@gmail.com> Co-authored-by: hfadzxy <starmoon_zhang@163.com> Signed-off-by: wangxiaoxin (A) <w00664509@china.huawei.com>
momo609
pushed a commit
to momo609/vllm-ascend
that referenced
this pull request
Jun 4, 2025
…ct#1040) ### What this PR does / why we need it? This PR enable accuracy test for PR labeled with "*accuracy-test" and workflow_dispatch. Only one model test running for each type test to reduce excution time. - The dense test costs about `25mins` to complete (gsm8k 7mins, ~mmlu 3h24mins,~ cEval 18mins) - The vl test costs about `40mins` to complete In futute, we might consider enable all job test as nightly schedule job. Below is mainly changes: - the dense/vl accuracy test will be triggered by lableling `accuracy-test` and `ready-for-test` - the dense accuracy test will be triggered by lableling `dense-accuracy-test` and `ready-for-test` - the vl accuracy test will be triggered by lableling `vl-accuracy-test` and `ready-for-test` - accuracy test will also be triggered by workflow_dispatch - Support V1 and V0 for qwen and V0 for VL For PR test we also generate summary in test summary. ### Does this PR introduce _any_ user-facing change? No ### How was this patch tested? - CI passed with accuracy-test label - Preview: https://github.com/vllm-project/vllm-ascend/actions/runs/15407628722?pr=1040 Closes: vllm-project#953 --------- Signed-off-by: hfadzxy <starmoon_zhang@163.com> Signed-off-by: Yikun Jiang <yikunkero@gmail.com> Co-authored-by: hfadzxy <starmoon_zhang@163.com> Signed-off-by: wangxiaoxin (A) <w00664509@china.huawei.com>
chopper0126
pushed a commit
to chopper0126/vllm-ascend
that referenced
this pull request
Oct 16, 2025
…ct#1040) ### What this PR does / why we need it? This PR enable accuracy test for PR labeled with "*accuracy-test" and workflow_dispatch. Only one model test running for each type test to reduce excution time. - The dense test costs about `25mins` to complete (gsm8k 7mins, ~mmlu 3h24mins,~ cEval 18mins) - The vl test costs about `40mins` to complete In futute, we might consider enable all job test as nightly schedule job. Below is mainly changes: - the dense/vl accuracy test will be triggered by lableling `accuracy-test` and `ready-for-test` - the dense accuracy test will be triggered by lableling `dense-accuracy-test` and `ready-for-test` - the vl accuracy test will be triggered by lableling `vl-accuracy-test` and `ready-for-test` - accuracy test will also be triggered by workflow_dispatch - Support V1 and V0 for qwen and V0 for VL For PR test we also generate summary in test summary. ### Does this PR introduce _any_ user-facing change? No ### How was this patch tested? - CI passed with accuracy-test label - Preview: https://github.com/vllm-project/vllm-ascend/actions/runs/15407628722?pr=1040 Closes: vllm-project#953 --------- Signed-off-by: hfadzxy <starmoon_zhang@163.com> Signed-off-by: Yikun Jiang <yikunkero@gmail.com> Co-authored-by: hfadzxy <starmoon_zhang@163.com>
Angazenn
pushed a commit
to Angazenn/vllm-ascend
that referenced
this pull request
Oct 21, 2025
…ct#1040) ### What this PR does / why we need it? This PR enable accuracy test for PR labeled with "*accuracy-test" and workflow_dispatch. Only one model test running for each type test to reduce excution time. - The dense test costs about `25mins` to complete (gsm8k 7mins, ~mmlu 3h24mins,~ cEval 18mins) - The vl test costs about `40mins` to complete In futute, we might consider enable all job test as nightly schedule job. Below is mainly changes: - the dense/vl accuracy test will be triggered by lableling `accuracy-test` and `ready-for-test` - the dense accuracy test will be triggered by lableling `dense-accuracy-test` and `ready-for-test` - the vl accuracy test will be triggered by lableling `vl-accuracy-test` and `ready-for-test` - accuracy test will also be triggered by workflow_dispatch - Support V1 and V0 for qwen and V0 for VL For PR test we also generate summary in test summary. ### Does this PR introduce _any_ user-facing change? No ### How was this patch tested? - CI passed with accuracy-test label - Preview: https://github.com/vllm-project/vllm-ascend/actions/runs/15407628722?pr=1040 Closes: vllm-project#953 --------- Signed-off-by: hfadzxy <starmoon_zhang@163.com> Signed-off-by: Yikun Jiang <yikunkero@gmail.com> Co-authored-by: hfadzxy <starmoon_zhang@163.com>
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
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.
What this PR does / why we need it?
Fix accuracy test
Does this PR introduce any user-facing change?
No
How was this patch tested?
Manually