#  torch-gpu

**Repository Path**: damone/torch-gpu

## Basic Information

- **Project Name**:  torch-gpu
- **Description**: Torch GPU docker 环境的创建及快速使用
- **Primary Language**: Unknown
- **License**: Not specified
- **Default Branch**: master
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 0
- **Forks**: 1
- **Created**: 2021-12-04
- **Last Updated**: 2023-04-21

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

# Pytorch 快速使用的容器案例

1. 创建
```sh
bash ./create.sh
```

2. 进入容器
```sh
bash ./entry.sh
```
3. 启动容器
容器创建自带了自动重启功能，一般不需要
```sh
bash ./start.sh
```

4. 销毁容器
```sh
bash ./clear.sh
```

5. 运行测试
默认进入容器后进入conda 的base环境，该环境安装了需要的pytorch 软件
运行如下进行测试

```
(base) root@8d326a869c84:~/example# python test.py 
99 1994.056396484375
199 1330.2864990234375
299 888.8197021484375
399 595.0929565429688
499 399.5873107910156
599 269.404296875
699 182.68043518066406
799 124.8812026977539
899 86.34074401855469
999 60.628963470458984
1099 43.46638870239258
1199 32.00395202636719
1299 24.344066619873047
1399 19.222064971923828
1499 15.794866561889648
1599 13.500123977661133
1699 11.962587356567383
1799 10.931628227233887
1899 10.23979377746582
1999 9.775188446044922
Result: y = 0.01719123311340809 + 0.8311278820037842 x + -0.0029657729901373386 x^2 + -0.08968716859817505 x^3
```