EasyML.jl

A foolproof way of doing ML with GUI elements.
Popularity
51 Stars
Updated Last
5 Months Ago
Started In
January 2021

EasyML.jl

docs stable docs dev CI main CI dev codecov

This package allows to use machine learning in Julia through a graphical user interface.

NB! This is a beta version. Bugs and breaking changes should be expected.

The package will be updated to Julia 1.7 as soon as QML.jl is updated.

If you are interested in the package and would like to contribute then write here: a_ill@outlook.com.

Features

It is possible to:

  • Design a neural network
  • Train a neural network
  • Validate a neural network
  • Apply a neural network to new data

Classification, regression and segmentation on images are currently supported.

Flux.jl machine learning library is used under the hood.

Installation

Run ] add EasyML in REPL.

If fonts do not look correct then install this and this font.

Quick guide

EasyML is easy enough to figure out by yourself! Just run the following lines.

Add the package

using EasyML

Set up

change(global_options)

Design

change_classes()
design_model()

Train

change(data_preparation_options)
change(training_options)
get_urls_training()
get_urls_testing()
prepare_training_data()
prepare_testing_data()
results = train()
remove_training_data()
remove_testing_data()
remove_training_results()

Validate

change(validation_options)
get_urls_validation()
results = validate()
remove_validation_data()
remove_validation_results()

Apply

change(application_options)
change_output_options()
get_urls_application()
apply()
remove_application_data()

On reopening

load_model()
load_options()

Development

A plan for the project can be seen here.