dockipy 0.0.8

No project description provided

Dockipy

Table of Contents

• Installation
• What is dockipy?
• Configure environment
• License

---

Installation

pip install dockipy

What is dockipy?

dockipy is your ticket to productivity paradise, a simple yet mighty wrapper around the Docker command line interface. It's the friend who says, "Don't worry about the nitty-gritty, I got you!" Just lean back and let dockipy work its magic, it will mount you project in a container seamlessly. If it doesn't break then everything should be fine.

And here's the cherry on top: with dockipy, you can dive into a Python virtual environment inside your container. It's like a containerception—codeception—whatever you want to call it! Run your code with the ease, all thanks to dockipy.

But wait, there's a catch! Before you sail into the dockipy, make sure you have Docker installed. And if you're planning to use NVIDIA GPUs, brace yourself! Docker GPU driver installation is a mess, dockipy won't hold your hand through that chaos—consider yourself warned!

How to Use dockipy

Ready to embark on your container adventure? Here's how dockipy can make your coding dreams come true:

Run Your Code: Use dockipy to run your code in a container inside a Python virtual environment within another container. It's like Inception, but for coding!

dockipy my_script.py

How to Use dockishell

Need to run an arbitrary command in a container? No problem! Just fire up dockishell and let the container magic begin.

dockishell nvidia-smi

How to Use dockinotebook

Want to unleash the power of Jupyter Notebooks in a container? Say no more! Use dockinotebook and open the link in your browser, colab or visual studio code.

dockinotebook

Configure environment

Tired of Docker feeling like a high-maintenance diva? Set up your environment like a pro with docki.yaml. No more Docker dramas, just smooth sailing.

docki --init

docki.yaml

This file is used to specify the base image, system dependencies, and python dependencies for the Docker container. If the file does not exist, a template will be created in the project root using docki --init.

base_image: The base image for the Docker container. You can find images on Docker Hub.

system_dep: A list of system dependencies to install in the Docker container. Install with apt-get.

python_dep: A list of python dependencies to install in the Docker container or a path to a requirements.txt file.

base_image: nvidia/cuda:11.8.0-devel-ubuntu22.04
shm_size: 16G # shared memory size
tag: docki_image
system_dep:
    - python3
    - python3-pip
    - python3-dev
    - python3-venv
python_dep:
    - jupyter
notebook_token: docki
notebook_password: docki

or

base_image: ubuntu:latest
shm_size: 16G # shared memory size
tag: docki_image
system_dep:
    - python3
    - python3-pip
    - python3-dev
    - python3-venv
python_dep:
    file: ./requirements.txt
notebook_token: docki
notebook_password: docki

TODO

• Add support Windows Notebooks
• Add support for Docker GPU driver installation on WLS2

License

dockipy is like that cool friend who always lets you borrow their stuff without any fuss. It's distributed under the terms of the apache-2.0 license. So go ahead, have some fun with it! 🎉