Skip to content
CI

chore(公开交付): 新.gitignore下全量研究数据一次性入库 #14

Sign in to view logs

chore(公开交付): 新.gitignore下全量研究数据一次性入库

chore(公开交付): 新.gitignore下全量研究数据一次性入库 #14

Workflow file for this run

.github/workflows/ci.yml at 6d3800c
name: CI
on:
push:
branches: [main, master]
pull_request:
branches: [main, master]
jobs:
validate:
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v4
- name: Set up Python 3.12
uses: actions/setup-python@v5
with:
python-version: "3.12"
- name: Install dependencies
run: |
python -m pip install --upgrade pip
pip install openai httpx anthropic
- name: Syntax check (benchmark/eval)
run: |
python -m py_compile benchmark/eval/validate_dataset.py
python -m py_compile benchmark/eval/evaluate.py
python -m py_compile benchmark/eval/judge.py
python -m py_compile benchmark/eval/utils.py
python -m py_compile benchmark/eval/report.py
python -m py_compile benchmark/eval/cross_domain_test.py
python -m py_compile benchmark/eval/cross_domain_radar.py
python -m py_compile benchmark/eval/cross_judge.py
python -m py_compile benchmark/eval/quality_audit.py
python -m py_compile benchmark/eval/self_correction_test.py
python -m py_compile benchmark/eval/self_correction_test_mimo.py
python -m py_compile benchmark/eval/self_correction_test_minimax.py
python -m py_compile benchmark/eval/summarize_multi.py
python -m py_compile benchmark/eval/t2_deep_analysis.py
python -m py_compile benchmark/eval/embedding_analysis.py
python -m py_compile benchmark/eval/dimension_waterfall.py
python -m py_compile benchmark/eval/formula_density.py
python -m py_compile benchmark/eval/image_formula_cooccurrence.py
python -m py_compile benchmark/eval/failure_analysis.py
python -m py_compile benchmark/eval/report_viz.py
python -m py_compile benchmark/eval/build_mimo_candidates.py
python -m py_compile benchmark/eval/build_minimax_candidates.py
python -m py_compile benchmark/eval/build_multimodal_index.py
python -m py_compile benchmark/eval/ping_models.py
python -m py_compile benchmark/eval/run_eval_pipeline.py
python -m py_compile benchmark/eval/monitor_eval.py
python -m py_compile benchmark/eval/check_leaderboard.py
- name: Syntax check (benchmark/pipeline)
run: |
python -m py_compile benchmark/pipeline/build_benchmark.py
python -m py_compile benchmark/pipeline/export_dataset.py
python -m py_compile benchmark/pipeline/merge_benchmarks.py
python -m py_compile benchmark/pipeline/split_benchmark.py
python -m py_compile benchmark/pipeline/validate_benchmark.py
python -m py_compile benchmark/pipeline/arbiter.py
python -m py_compile benchmark/pipeline/orphan_judge.py
python -m py_compile benchmark/pipeline/compare_mineru_vs_pymupdf.py
- name: Syntax check (benchmark/generator)
run: |
python -m py_compile benchmark/generator/api.py
python -m py_compile benchmark/generator/prompts.py
python -m py_compile benchmark/generator/distribution.py
python -m py_compile benchmark/generator/quality.py
- name: Validate dataset (core.json)
run: python benchmark/eval/validate_dataset.py --input benchmark/dataset/core.json
- name: Reference evaluation (core.json)
run: python benchmark/eval/evaluate.py --mode reference --input benchmark/dataset/core.json --output /tmp/core_report.json
- name: Reference evaluation (sample)
run: python benchmark/eval/evaluate.py --mode reference --input benchmark/dataset/sample_questions.json --output /tmp/sample_report.json