# mindspore_lite_samples

**Repository Path**: don-quixote-der/mindspore_lite_samples

## Basic Information

- **Project Name**: mindspore_lite_samples
- **Description**: No description available
- **Primary Language**: Unknown
- **License**: Apache-2.0
- **Default Branch**: master
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 0
- **Forks**: 0
- **Created**: 2023-11-02
- **Last Updated**: 2023-11-02

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

<center font='bold'><h1>Mindspore_lite安装使用说明</h1></center>

## 安装

在这下载：https://www.mindspore.cn/lite/docs/zh-CN/r2.2/use/downloads.html

在昇腾环境上 注意mindspore_lite与CANN的对应关系

### C++安装

下载并解压 C++，需要配置LD_LIBRARY_PATH

```shell
wget https://ms-release.obs.cn-north-4.myhuaweicloud.com/2.2.0/MindSpore/lite/release/linux/aarch64/cloud_fusion/python37/mindspore-lite-2.2.0-linux-aarch64.tar.gz
tar -zxvf mindspore-lite-2.2.0-linux-aarch64.tar.gz
```

### Python安装

下载python使用包

注意下载对应CPU架构与python版本

```shell
wget https://ms-release.obs.cn-north-4.myhuaweicloud.com/2.2.0/MindSpore/lite/release/linux/aarch64/cloud_fusion/python39/mindspore_lite-2.2.0-cp39-cp39-linux_aarch64.whl

pip install mindspore_lite-2.2.0-cp39-cp39-linux_aarch64.whl
```


## 推理

### Python

快速入门例子

云端推理参考 https://www.mindspore.cn/lite/docs/zh-CN/r2.2/use/cloud_infer/runtime_python.html

python快速入门（CPU推理）

下载模型与数据

```shell
mkdir model

wget https://download.mindspore.cn/model_zoo/official/lite/quick_start/mobilenetv2.mindir

wget  https://download.mindspore.cn/model_zoo/official/lite/quick_start/input.bin
```

执行quick_demo下的	quick_start_cloud_infer_python.py

python quick_start_cloud_infer_python.py

### 转换ONNX demo

下载转换工具

wget https://ms-release.obs.cn-north-4.myhuaweicloud.com/2.0.0/MindSpore/lite/release/linux/aarch64/cloud_fusion/python37/mindspore-lite-2.0.0-linux-aarch64.tar.gz

tar -zxvf mindspore-lite-2.0.0-linux-aarch64.tar.gz

将转换工具需要的动态链接库加入环境变量LD_LIBRARY_PATH。

```
export PACKAGE_ROOT_PATH=/home/ma-user/work/mindspore-lite-2.0.0-linux-aarch64
export LD_LIBRARY_PATH=${PACKAGE_ROOT_PATH}/tools/converter/lib:${LD_LIBRARY_PATH}
export PATH=${PACKAGE_ROOT_PATH}/tools/converter/converter:$PATH
```

转换模型

```
converter_lite --fmk=ONNX --saveType=MINDIR --optimize=none --modelFile=safeHelmet.onnx --outputFile=safeHelmet
```

提示转换成功

后执行代码中的msLite_infer.py执行即可





## 精度



todo

## 性能

todo