deepNeuralNetwork.run: Partial Prediction for in-training use.

In OscarGVelasco/DeepNeuralNetworks4R: Implementation of Deep Neural Networks in R language

Description

'deepNeuralNetwork.run()' returns a partial prediction over the test data to measure overfitting if test data is supplied by the user. It is calculated using the actual in-training-model.

Usage

deepNeuralNetwork.run(model.trained, data = X.test)

Arguments

model.trained	An object of class DeepNNModel.
data	Matrix containing the test data without the observed variable that we want to predict.

Details

This function will calculate a prediction using test data (if it is supplied by the user). [deepNeuralNetwork.training] function will use this function and depending on the result given it can decide whether to continue with the training or stop it.

Value

A vector or matrix containing the prediction values.

OscarGVelasco/DeepNeuralNetworks4R documentation built on Jan. 24, 2021, 12:42 a.m.