# Prostate_DL

**Repository Path**: human-robot/Prostate_DL

## Basic Information

- **Project Name**: Prostate_DL
- **Description**: No description available
- **Primary Language**: Unknown
- **License**: Not specified
- **Default Branch**: master
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 0
- **Forks**: 0
- **Created**: 2018-02-28
- **Last Updated**: 2020-12-18

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

# Prostate_DL

如何在docker 中训练
1. 创建镜像或者使用已有的tensorflow 镜像。 
`docker images`
-  tenserflow/prostate 安装了git , pip , opencslide 专门训练wsi图片。
-  tensorflow/tensorflow 没有安装其他的图片处理包。
-  输出
REPOSITORY              TAG                    IMAGE ID            CREATED             SIZE
tenserflow/prostate     latest                 d5d25dbe2c44        6 seconds ago       4.54GB
tensorflow/tensorflow   latest-gpu             e93ab8001bbf        4 weeks ago         2.95GB


2. 运行docker，挂载数据集 volume， 并执行jupyter notebook 
`nvidia-docker run -v /home/ramsley/DataSet/prostate:/DataSet/prostate -it -p 9999:8888 tenserflow/prostate /run_jupyter.sh --allow-root`
- 输出
[I 01:12:16.216 NotebookApp] http://[all ip addresses on your system]:8888/?token=3084c932e66272d122c8efcc25d84e05c3bd53921e2f984b
[I 01:12:16.216 NotebookApp] Use Control-C to stop this server and shut down all kernels (twice to skip confirmation).
[C 01:12:16.216 NotebookApp]
    
    Copy/paste this URL into your browser when you connect for the first time,
    to login with a token:
        http://localhost:8888/?token=3084c932e66272d122c8efcc25d84e05c3bd53921e2f984b


3. 打开浏览器输入 http://localhost:8888/?token=3084c932e66272d122c8efcc25d84e05c3bd53921e2f984b  ， 修改端口号8888->9999.  localhost -> 主机IP。 


4. 在notebook中导航到 home/user/git/ 目录下使用已有的code训练， 或者将自己的code倒入到contanier中， 建议用git 来控制代码。
`git clone https://gitee.com/smartdraw/Prostate_DL.git `