mcnets 2.0.1

Various machine learning models that use unique Monte Carlo-type parameter training.

Monte-Carlo-Neural-Nets

Overview

Originally a hobby project, this package now represents neural network models that train using a Monte-Carlo algorithm and are on par with other models such as the MLPRegressor in SciKit-Learn.

Currently in V2.0.0, the list of avaliable models is only the primary NeuralNetwork class. In future updates, there will be additional models that are built on top of this base, or are revised classes from the previous main version 1.5.0.

Some primary features include:

• Neural Network model with full customization support
• Various built-in activation functions (with support for external ones)
• Built-in test-train-split (TTSplit) and cross-validation (cross_val) functions
• Built-in scorer functions with support for external functions
• Ability to save and load models at any point

GitHub and QuickStart

More explanations, examples, and technicals can be found on the GitHub page: https://github.com/SciCapt/Monte-Carlo-Neural-Nets