ML Template

A sample repo for structuring ML projects

Setting up your local environment (in windows)

• Make sure to have chocolatey installed
• Install Docker using choco install docker-desktop
• Install make using choco install make
• Have conda/anaconda installed
• run make env within the root directory of this project to create an empty environment
• Activate environment with conda activate ml_template
• install all dependencies by running make deps from the root directory of this project

Making and merging branches

• run git checkout main, this will get you the main version of the code
• run git pull origin main, this will grab the latest main version from github.com
• run git checkout -b <your_name>/<issue_number> to create a new branch
• Make you changes to the code
• run git add <file_name> to add say you want to keep the changes
• run git commit -m "<issue_number> : decription" to create a snapshot of your updated code
• run git push origin <branch_name> to upload your code to github.com

Running tests

• Run make tests to run unit test suite in the ./unittests folder using pytest

Using docker as a dev environment