Metadata-Version: 2.4
Name: pae-toolkit
Version: 0.0.1.20250909
Summary: PAE toolkit
Author-email: Yanguan02 <yanguan_02@126.com>
Requires-Python: >=3.10
Requires-Dist: ffmpeg-python>=0.2.0
Requires-Dist: librosa>=0.11.0
Requires-Dist: onnx>=1.19.0
Requires-Dist: onnxruntime>=1.22.1
Requires-Dist: onnxsim>=0.4.36
Requires-Dist: opencv-python-headless>=4.12.0.88
Requires-Dist: pydantic>=2.11.5
Requires-Dist: pytest>=8.4.1
Requires-Dist: regex>=2024.11.6
Requires-Dist: requests>=2.32.4
Requires-Dist: torch==2.7.1
Requires-Dist: torchaudio==2.7.1
Requires-Dist: transformers>=4.56.1
Requires-Dist: typer>=0.16.0
Description-Content-Type: text/markdown

# PAE toolkit

## Description

- [msprechecker]()

- [modelscope](https://www.modelscope.cn/docs/models/download) is used for downloading models when can't download from huggingface

- [evalscope](https://evalscope.readthedocs.io/zh-cn/latest/user_guides/stress_test/quick_start.html) is used for evaluating the Inference Engine of Large Language Models

## Install

```bash
pip install pae-toolkit -i https://pypi.org/simple
```

```bash
pip install modelscope
pip install evalscope[perf]
pip install msprechecker
```

## Usage

- `pae_toolkit check-mindie-high-preformance-env`


```bash
git submodule add https://gitcode.com/Ascend/msit.git third_party/msit
```