Multi-cloud setup

Code to setup working environment for DS/ML/AI projects on the cloud.

Tested with AWS, GCP & Azure and is what I regularly use to set up cloud environments before I begin work on a project. Uses conda for environment management and pip/conda for package management.

Usage

Clone this repo:

git clone https://github.com/gradhakr/multicloud_setup.git

Specific parts

Miniconda

To install the latest Linux version of miniconda from the Anaconda website, do:

make conda

Default DS environment

To setup a working conda environment with popular DS libraries (pandas, numpy, sklearn, seaborn, plotly, pytest, do:

conda env create -f .env_requirements/ds_env.yml

A minimal test

You can run a quick test to test your python install using:

make test