Wolfram Cloud Resource Object

Paclet Resource

AntonAntonov/MonadicContextualClassification

A monad for machine learning classification workflows

Contributed By: Anton Antonov

A software monad for Machine Learning (ML) classification workflows. Streamlined data onboarding. Both built-in and ensemble classifiers are supported. Receiver Operating Characteristic (ROC) functions can be specified, computed, and plotted. Other built-in classifier measurements and ML classification operations are implemented and supported.

Installation Instructions

To install this paclet in your Wolfram Language environment, evaluate this code:
PacletInstall[ResourceObject["https://wolfr.am/1ozsJYvwM"]]


To load the code after installation, evaluate this code:
Needs["AntonAntonov`MonadicContextualClassification`"]

Details

The monad facilitates the rapid specification of classification workflows.
A (short) workflow is: split the data, make a classifier, compute and show classifier measurements, show Receiver Operating Characteristic (ROC) plots.
If the given data is a dataset, then by default the last column is considered to have the class labels.
The splitting utilizes random sampling.
By default class-label stratification is used for the data splitting.
Both built-in classifiers and ensemble classifiers can be used in monads pipelines.
The computation of the (standard) built-in classifier measurements is supported.
Special monad functions are used for the ROC plots.

Examples

Disclosures

  • Paclet dependencies

Paclet Source

Source Metadata

  • Citation: {TextData[ButtonBox["A monad for classification workflows | Mathematica for prediction algorithms", BaseStyle -> "Hyperlink", ButtonData -> {URL["https://mathematicaforprediction.wordpress.com/2018/05/15/a-monad-for-classification-workflows/"], None}, ButtonNote -> "https://mathematicaforprediction.wordpress.com/2018/05/15/a-monad-for-classification-workflows/"]]}

See Also