WALEX2000 / qml

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

QML

Qml is a CLI tool that enables AI Engineers to create and manage simple Automated ML Pipelines, for experimental or small-scope local ML projects.

Setup

To install for prod. run: rm -rf build dist *.egg-info (To remove previous build) python setup.py sdist bdist_wheel (To build) pipx install . --force (To install) To install for dev. run: source .venv/bin/activate then, just call qml (It's already on the path)

qml is a tool to quickstart machine learning projects, while supporting lightweight MLOps tasks

Folder Structure

To create a new Pipeline Environment Configuration file, add it to the root of 'qml_assets' directory. Make sure to make the property 'name' the same as the file name (without the initial . and the .yaml extension)

Create a sub-directory inside qml_assets with the following name -assets. Here you'll place all assets to be used inside your project

In the modules directory, create a sub-directory with the following name: -modules All modules (commands, event processes, and setup processes), should be placed inside this new folder.

About

License:MIT License


Languages

Language:Python 99.1%Language:Dockerfile 0.9%