HiCiChlid / geosci-pyenv

An environment for geo-science based on docker

geosci-pyenv

Contents:

1. python3
2. R
3. Spark
4. QGIS

How to install it:

Method 1

1. Download docker and install it (https://www.docker.com/get-started).

2. Start docker using systemctl start docker.service in Linux; or sudo service docker start in WSL2 linux-sub-system in Windows; or double click docker icons to make sure it open in Windows/MacOS.

3. RUN

   docker build -t [your_image_name]:[version] https://raw.githubusercontent.com/HiCiChlid/geosci-pyenv/main/dockerfile

4. Wait for it to finish.

---

Method 2

1. Download docker and install it (https://www.docker.com/get-started).

2. Start docker using systemctl start docker.service in Linux; or sudo service docker start in WSL2 linux-sub-system in Windows; or double click docker icons to make sure it open in Windows/MacOS.

3. Download dockerfile (https://github.com/HiCiChlid/geosci-pyenv/blob/main/dockerfile) to the specified location.

4. Open Terminal or CMD, and locate to the specified location using cd (in Windows for switching to another volume, you can directly input the volume mark in the CMD, e.g., D:).

5. Run

   docker build -t [your_image_name]:[version] [dockerfile location], e.g., docker build -t test:v1 .

6. Wait for it to finish.

How to use it:

1. Open Terminal and make sure docker is open.

2. Run

   docker run -v [connected_path_in_your_computer]:/home/current -id --name=[container_name] -p 8888:8888 -p 4040:4040 --shm-size 2g [your_image_name]:[version]

   Note that: in windows, [connected_path_in_your_computer] is in the format of /mnt/d/"path with space/next layer/next layer"

3. In windows, if you hope to exploit your Nvidia GUP in docker, you can download and install the driver first (https://developer.nvidia.com/cuda/wsl/download), and then run

   docker run -v [connected_path_in_your_computer]:/home/current -id --name=[container_name] -p 8888:8888 -p 4040:4040 --shm-size 2g --gpus all [your_image_name]:[version]

4. After container building, you can start your container through

   docker exec -it [container_name] /bin/bash

5. you can use sh jp.sh to start Jupyter notebook.

How to use VScode to debug Python in Windows WSL sub-linux system:

1. Download VScode(https://code.visualstudio.com/download) and install it.
2. Search Remote - Containers, Remote - WSL, Docker from Extension Marketplace.

3. Click the Docker icon and start your container

4. Click the green bar in the bottom

,

and choose Attach to running container

and then choose the container with python3, and a new window will open.

5. In new the new window, click the extension

to search Python and install it. You can enjoy debuging your .py scripts. :)