There is a lot hack code for v0.11.0, which makes the code hard to
upgrade to newer vLLM version. Since v0.11.0 will release soon. Let's
drop v0.11.0 support first. Then we'll upgrade to v0.11.2 soon.
- vLLM version: v0.11.0
- vLLM main:
2918c1b49c
Signed-off-by: wangxiyuan <wangxiyuan1007@gmail.com>
173 lines
6.1 KiB
YAML
173 lines
6.1 KiB
YAML
#
|
|
# Copyright (c) 2025 Huawei Technologies Co., Ltd. All Rights Reserved.
|
|
#
|
|
# Licensed under the Apache License, Version 2.0 (the "License");
|
|
# you may not use this file except in compliance with the License.
|
|
# You may obtain a copy of the License at
|
|
#
|
|
# http://www.apache.org/licenses/LICENSE-2.0
|
|
#
|
|
# Unless required by applicable law or agreed to in writing, software
|
|
# distributed under the License is distributed on an "AS IS" BASIS,
|
|
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
|
# See the License for the specific language governing permissions and
|
|
# limitations under the License.
|
|
# This file is a part of the vllm-ascend project.
|
|
#
|
|
|
|
# This test will be triggered:
|
|
# 1. schedule
|
|
# 2. pull_request change the related files
|
|
# 3. workflow_dispatch with models input
|
|
|
|
name: ascend test / accuracy report
|
|
|
|
on:
|
|
pull_request:
|
|
branches:
|
|
- 'main'
|
|
- '*-dev'
|
|
paths:
|
|
- '.github/workflows/vllm_ascend_test_report.yaml'
|
|
- 'tests/e2e/models/test_lm_eval_correctness.py'
|
|
workflow_dispatch:
|
|
inputs:
|
|
vllm-ascend-version:
|
|
description: 'vllm-ascend:'
|
|
required: true
|
|
type: choice
|
|
# Current supported vLLM versions
|
|
options:
|
|
- latest
|
|
- main
|
|
default: main
|
|
|
|
# Bash shells do not use ~/.profile or ~/.bashrc so these shells need to be explicitly
|
|
# declared as "shell: bash -el {0}" on steps that need to be properly activated.
|
|
# It's used to activate ascend-toolkit environment variables.
|
|
defaults:
|
|
run:
|
|
shell: bash -el {0}
|
|
|
|
# only cancel in-progress runs of the same workflow
|
|
concurrency:
|
|
group: ${{ github.workflow }}-${{ github.ref }}
|
|
cancel-in-progress: true
|
|
|
|
jobs:
|
|
run:
|
|
strategy:
|
|
fail-fast: false
|
|
matrix:
|
|
include:
|
|
- runner: linux-aarch64-a2-1
|
|
model_list:
|
|
- Qwen3-8B
|
|
- Qwen2.5-VL-7B-Instruct
|
|
- Qwen2-Audio-7B-Instruct
|
|
- runner: linux-aarch64-a2-2
|
|
model_list:
|
|
- Qwen3-30B-A3B
|
|
- Qwen3-VL-30B-A3B-Instruct
|
|
- DeepSeek-V2-Lite
|
|
uses: ./.github/workflows/_e2e_nightly_single_node_models.yaml
|
|
with:
|
|
vllm: 2918c1b49c88c29783c86f78d2c4221cb9622379
|
|
runner: ${{ matrix.runner }}
|
|
image: swr.cn-southwest-2.myhuaweicloud.com/base_image/ascend-ci/cann:8.3.rc1-910b-ubuntu22.04-py3.11
|
|
model_list: ${{ toJson(matrix.model_list) }}
|
|
upload: ${{ github.event_name == 'workflow_dispatch' && github.event.inputs.vllm-ascend-version == 'latest' }}
|
|
|
|
create_pr:
|
|
runs-on: ubuntu-latest
|
|
needs: run
|
|
if: ${{ github.event_name == 'workflow_dispatch' && github.event.inputs.vllm-ascend-version == 'latest' }}
|
|
env:
|
|
UPSTREAM_REPO: vllm-project/vllm-ascend
|
|
steps:
|
|
- name: Checkout repository
|
|
uses: actions/checkout@v4
|
|
with:
|
|
repository: vllm-ascend-ci/vllm-ascend
|
|
token: ${{ secrets.PAT_TOKEN }}
|
|
ref: main
|
|
|
|
- name: Add upstream remote
|
|
run: |
|
|
git remote add upstream https://github.com/${{ env.UPSTREAM_REPO }}.git
|
|
git fetch upstream
|
|
git remote -v
|
|
|
|
- name: Set Git user info dynamically
|
|
run: |
|
|
git config user.name "${{ github.actor }}"
|
|
git config user.email "${{ github.actor }}@users.noreply.github.com"
|
|
|
|
- name: Create or switch to branch
|
|
run: |
|
|
TIMESTAMP=$(date +%Y%m%d%H%M%S)
|
|
BRANCH_NAME="auto-pr/accuracy-report-${TIMESTAMP}"
|
|
echo "BRANCH_NAME=${BRANCH_NAME}" >> $GITHUB_ENV
|
|
git checkout -B "${BRANCH_NAME}" upstream/main
|
|
|
|
- name: Download only current run reports
|
|
uses: actions/download-artifact@v6
|
|
with:
|
|
path: ./docs/source/developer_guide/evaluation/accuracy_report
|
|
pattern: report-*
|
|
github-token: ${{ secrets.GITHUB_TOKEN }}
|
|
run-id: ${{ github.run_id }}
|
|
|
|
- name: Delete old report
|
|
run: |
|
|
find ./docs/source/developer_guide/evaluation/accuracy_report -maxdepth 1 -type f -name '*.md' ! -name 'index.md' -delete
|
|
find ./docs/source/developer_guide/evaluation/accuracy_report -mindepth 2 -type f -name '*.md' -exec mv -f {} ./docs/source/developer_guide/evaluation/accuracy_report \;
|
|
find ./docs/source/developer_guide/evaluation/accuracy_report -mindepth 1 -type d -empty -delete
|
|
|
|
- name: Update accuracy_report/index.md
|
|
run: |
|
|
REPORT_DIR="./docs/source/developer_guide/evaluation/accuracy_report"
|
|
INDEX_MD="$REPORT_DIR/index.md"
|
|
{
|
|
echo "# Accuracy Report"
|
|
echo ""
|
|
echo ":::{toctree}"
|
|
echo ":caption: Accuracy Report"
|
|
echo ":maxdepth: 1"
|
|
|
|
for report in "$REPORT_DIR"/*.md; do
|
|
filename="$(basename "$report" .md)"
|
|
if [ "$filename" != "index" ]; then
|
|
echo "$filename"
|
|
fi
|
|
done
|
|
echo ":::"
|
|
} > "$INDEX_MD"
|
|
|
|
- name: push accuracy report
|
|
env:
|
|
GITHUB_TOKEN: ${{ secrets.PAT_TOKEN }}
|
|
run: |
|
|
git add ./docs/source/developer_guide/evaluation/accuracy_report/*.md
|
|
git commit -s -m "[Doc] Update accuracy reports for ${{ env.BRANCH_NAME }}"
|
|
git push -f origin "${{ env.BRANCH_NAME }}"
|
|
|
|
- name: Create PR in upstream via API
|
|
uses: actions/github-script@v8
|
|
with:
|
|
github-token: ${{ secrets.PAT_TOKEN }}
|
|
script: |
|
|
const pr = await github.rest.pulls.create({
|
|
owner: 'vllm-project',
|
|
repo: 'vllm-ascend',
|
|
head: `vllm-ascend-ci:${{ env.BRANCH_NAME }}`,
|
|
base: 'main',
|
|
title: `[Doc] Update accuracy reports for ${{ env.BRANCH_NAME }}`,
|
|
body: `The accuracy results running on NPU Altlas A2 have changed, updating reports for: All models
|
|
|
|
- [Workflow run][1]
|
|
|
|
[1]: ${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }}`
|
|
});
|
|
core.info(`Created PR #${pr.data.number}`);
|