Skip to content

Qiushao-E/MSR-Task

Repository files navigation

Qwen2.5-Coder-0.5B-Instruct HumanEval 评估

环境设置

  1. 安装依赖:
uv sync
  1. 启动模型服务:

需要docker环境启动服务,使用vllm/vllm-openai:latest镜像

uv run python scripts/serve_model.py

或者在本地启动服务

uv run python scripts/serve_model_local.py

服务将在 http://localhost:3344 启动

测试服务

curl http://localhost:3344/v1/models

运行推理

uv run python scripts/inference.py

结果保存在 outputs/humaneval_results.jsonl

运行评估

uv run python scripts/evaluate.py

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors

Languages