### What this PR does / why we need it?
This PR enabled pytest and yaml style accuracy test, users now can
enable accuracy test by running:
```bash
cd ~/vllm-ascend
pytest -sv ./tests/e2e/singlecard/models/test_lm_eval_correctness.py \
--config ./tests/e2e/singlecard/models/configs/Qwen3-8B-Base.yaml \
--report_output ./benchmarks/accuracy/Qwen3-8B-Base.md
pytest -sv ./tests/e2e/singlecard/models/test_lm_eval_correctness.py \
--config-list-file ./tests/e2e/singlecard/models/configs/accuracy.txt
```
Closes: https://github.com/vllm-project/vllm-ascend/issues/1970
### Does this PR introduce _any_ user-facing change?
no
### How was this patch tested?
- vLLM version: v0.10.0
- vLLM main:
2836dd73f1
---------
Signed-off-by: Icey <1790571317@qq.com>
1.2 KiB
1.2 KiB
{{ model_name }}
vLLM Version: vLLM: {{ vllm_version }} ([{{ vllm_commit[:7] }}](https://github.com/vllm-project/vllm/commit/{{ vllm_commit }})),
vLLM Ascend Version: {{ vllm_ascend_version }} ([{{ vllm_ascend_commit[:7] }}](https://github.com/vllm-project/vllm-ascend/commit/{{ vllm_ascend_commit }}))
Software Environment: CANN: {{ cann_version }}, PyTorch: {{ torch_version }}, torch-npu: {{ torch_npu_version }}
Hardware Environment: Atlas A2 Series
Datasets: {{ datasets }}
Parallel Mode: TP
Execution Mode: ACLGraph
Command:
export MODEL_ARGS={{ model_args }}
lm_eval --model {{ model_type }} --model_args $MODEL_ARGS --tasks {{ datasets }} \
--apply_chat_template {{ apply_chat_template }} --fewshot_as_multiturn {{ fewshot_as_multiturn }} {% if num_fewshot is defined and num_fewshot != "N/A" %} --num_fewshot {{ num_fewshot }} {% endif %} \
--limit {{ limit }} --batch_size {{ batch_size}}
| Task | Metric | Value | Stderr |
|---|---|---|---|
| {% for row in rows -%} | |||
| {{ row.task.rjust(23) }} | {{ row.metric.rjust(15) }} | {{ row.value }} | ± {{ "%.4f" |
| {% endfor %} |