How to Package A Python Application using Anaconda and Docker

Quick tutorial for packaging Python project using Anaconda with Docker

I write this quick tutorial for those who want to create a python app with anaconda packages and package the app using Docker.

A project to quickly setup a python project using anaconda packages and Docker GitHub is home to over 50 million… github.com

Project Structure

.
├── Dockerfile
├── environment.yaml
├── main.py
├── README.md
└── requirements.txt

Create a YAML file for Conda Environment

environment.txt

name: app

channels:
  - conda-forge
  - anaconda

dependencies:
  - flask
  - gunicorn
  - pandas
  - matplotlib
  - scikit-learn
  - pip:
      - -r file:requirements.txt

• Name your conda environment, for instance app
• Add a list of channels where the desired packages are hosted
• List packages you want to install such as flask, gunicorn, numpy… under dependencies
• You can also install pip packages from the YAML file. One way to it is by adding a pip command to install (pip) packages from a requirements.txt, (-r in the command stands for requirement)

For instance:

requirements.txt

numpy

Create a Python Script

Here we have created a test file main.py for testing the application.

main.py

import numpy as np


print('Hello World !')
print('Numpy version is ', np.__version__)

Dokerization

Create a file named Dockerfile (without extension) from which we build a docker image for our Python application.

Dockerfile

FROM  continuumio/miniconda3
LABEL Author, Amine HadjYoucef

ENV APP_HOME /app
WORKDIR $APP_HOME
COPY . $APP_HOME

#---------------- Prepare the envirennment
RUN conda update --name base conda &&\
    conda env create --file environment.yaml
SHELL ["conda", "run", "--name", "app", "/bin/bash", "-c"]

ENTRYPOINT ["conda", "run", "--name", "app", "python", "main.py"]

• For simplicity we start from the base image * of miniconda3 (line1)
• Copy the files from the folder to docker * (line 6)
• Create a conda environment and install the packages defined in YAML file (line 9)
• Edit the shell command so we can launch python scripts from within the newly created conda environment (line 11)
• Define an entry-point for the docker image such that it execute the main.py python script of the application (line 13)

Build a docker image

We build a docker image, named app, with the following commands

docker build --tag myapp .

… then wait for docker to build an image for your app (it might tight a take for downloading and update it the packages list).

Note:

The Dockerfile must be located in the root directory of the application
Make sure Docker is installed on your machine (www.docker.com).

Run the docker image

To run the App from the docker image we created

docker run --rm -ti myapp

This should outputs:

Hello World !
Numpy version is  1.19.4

There you have it.

PS: If you encounter a problem please create an issue on GitHub or comment below. Resources