-
Notifications
You must be signed in to change notification settings - Fork 46
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
add Proposal for Large Language Model Edge Benchmark Suite: Implement… #127
Conversation
…ation on KubeEdge-lanvs Signed-off-by: yexiaochuan <tap91624@gmail.com>
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Overall looks great to me. A few suggestions:
- The scheme is not purely of a single task. It would more like to be a compression version of the single task.
- GPU environment is necessary and a good point. But the current design is not clear. The overall Hardware configuration should be compatible with the previous version of yaml files.
- Review the system metrics, especially the inference phase.
- Highlight the modified part of the architecture.
@hsj576 might also need to take a look at this proposal |
Overall looks great to me. Just a few questions: |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
/lgtm |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
/lgtm
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
The implementation has been removed, which follows the suggestion made in the routine meeting. Overall it looks fine to me.
But some tiny concern
- the documents are a little bit confusing: why there are "llm-benchmark-suit" and "llm-benchmark-suite"?
- besides, the "edge" shall be highlighted in the name of the documents
the benchmark suite description was fixed and the edge was highlighted in the nameing of documents 4244a71 |
…ation on KubeEdge-lanvs Signed-off-by: yexiaochuan <yxc2020@foxmail.com>
Signed-off-by: yexiaochuan <yxc2020@foxmail.com>
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Looks fine to me for the revised version. As soon as the CI issues are fixed, it would be fine for ianvs to merge this proposal.
Run pylint '/home/runner/work/ianvs/ianvs/core'
************* Module core.testenvmanager.dataset.dataset
core/testenvmanager/dataset/dataset.py:119:0: W0012: Unknown option value for 'disable', expected a valid pylint message and got 'too-many-positional-arguments' (unknown-option-value)
core/testenvmanager/dataset/dataset.py:206:0: C0303: Trailing whitespace (trailing-whitespace)
core/testenvmanager/dataset/dataset.py:214:0: C0303: Trailing whitespace (trailing-whitespace)
-----------------------------------
Your code has been rated at 9.9[8](https://github.com/kubeedge/ianvs/actions/runs/11338050517/job/31531990780#step:5:9)/10
Error: The operation was canceled.
This reverts commit f341b0f. Signed-off-by: yexiaochuan <yxc2020@foxmail.com>
…me of the documents Signed-off-by: yexiaochuan <yxc2020@foxmail.com>
Signed-off-by: yexiaochuan <yxc2020@foxmail.com>
Signed-off-by: yexiaochuan <yxc2020@foxmail.com>
Signed-off-by: yexiaochuan <yxc2020@foxmail.com>
Signed-off-by: yexiaochuan <yxc2020@foxmail.com>
Seems that pylint message is not valid (as below). @XueSongTap might want to try out https://stackoverflow.com/questions/79019204/too-many-positional-arguments-on-one-machine-but-does-not-know-the-error-on-t
|
Signed-off-by: yexiaochuan <yxc2020@foxmail.com>
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Need to fix below CI-error messages before merging.
Run pylint '/home/runner/work/ianvs/ianvs/core'
************* Module core.testenvmanager.dataset.dataset
core/testenvmanager/dataset/dataset.py:119:0: W0012: Unknown option value for 'disable', expected a valid pylint message and got 'R0917' (unknown-option-value) core/testenvmanager/dataset/dataset.py:207:0: W0012: Unknown option value for 'disable', expected a valid pylint message and got 'R0917' (unknown-option-value)
core/testenvmanager/dataset/dataset.py:215:0: W0012: Unknown option value for 'disable', expected a valid pylint message and got 'R0917' (unknown-option-value) core/testenvmanager/dataset/dataset.py:249:0: W0012: Unknown option value for 'disable', expected a valid pylint message and got 'R0917' (unknown-option-value)
core/testenvmanager/dataset/dataset.py:289:0: W0012: Unknown option value for 'disable', expected a valid pylint message and got 'R0917' (unknown-option-value) core/testenvmanager/dataset/dataset.py:374:0: W0012: Unknown option value for 'disable', expected a valid pylint message and got 'R0917' (unknown-option-value)
-----------------------------------
Your code has been rated at 9.96/10
Error: Process completed with exit code 4.
BTW: @XueSongTap might want to try out this solution to similar error messages, as mentioned above
Signed-off-by: yexiaochuan <yxc2020@foxmail.com>
Signed-off-by: yexiaochuan <yxc2020@foxmail.com>
Hi,I have resolved the pylint CI issues. Please review the changes and consider merging this PR. |
/lgtm |
/approve |
[APPROVALNOTIFIER] This PR is APPROVED This pull-request has been approved by: hsj576, MooreZheng The full list of commands accepted by this bot can be found here. The pull request process is described here
Needs approval from an approver in each of these files:
Approvers can indicate their approval by writing |
…ation on KubeEdge-lanvs
What type of PR is this?
/kind design
What this PR does / why we need it:
This pr is a proposal for Large Language Model Edge Benchmark Suite: Implementation on KubeEdge-lanvs
Which issue(s) this PR fixes:
Fixes # #94