nextai 1.0.0

A simple cli utility to generate ML project structure for quickly starting ML projects

Project Structure for MLflow integrated ML Projects

This cli tool generates the following directory structure for quickstart ML projects

installaton:

pip install nextai

example use:

nextai create mytestproject

It generates the project with following structure

Project
|
+--- Input
|    |
|    +--- raw               Raw data here
|    |
|    +--- interim           Any intermediate data, to pause and continue experiments
|    |
|    +--- processed         Processed data ready for ML pipeline
|
+--- output
|    |
|    +--- models             Model pickle or model weights stored here
|    |
|    +--- artifacts         Serialized artifacts like LabelEncoder, Vectorizer etc
|    |
|    +--- images            All plots and visualizations goes here
|    |
|    +--- Results           If the results needs to be stored for review, save here
|
+--- notebooks              All notebooks and experiments resides here
|
+--- src                    Final program, with training and prediction pipeline
|
|
README.md                   Description and instruction about the project
nextai                   MLflow project file. If you want to use this directory as MLflow project
Requirements.txt            python dependencies
Config.yml                  configuration key values in yaml format