Machine Learning JS is a node module designed to be a simple and easy to use library so that you don't have to know all of how training and optimization works to create a simple AI.

First, you need to have node and npm installed. Once you have that installed, go to the project directory and run npm install machinelearningjs.

I'll move this to a jsdoc eventually.

const ml = require('machinelearningjs')

To create a neural network, use const NETWORKNAME = ml.network(NUMINPUTS(int), NUMOUTPUTS(int), NUMHIDDENLAYERs(int)).

To train the neural network, use const ERRORS = NETWORKNAME.train(XTRAIN(arr), YTRAIN(arr), LOSS(func), LOSSPRIME(func), LEARNINGRATE(float), EPOCHS(int), VERBOSE(bool)). This returns a list of all the errors from each epoch.

To receive a prediction, use NETWORKNAME.predict(INPUT(arr)).