SM.rnn-class: Class SM.rnn - An S4 class contains a recurrent neural...

In 4301350/sm4sd: Statistical Modeling for Sensor Data

Description

SM.rnn class contains the trained rnn model and its parameters.

objects from the Class

Objects are created by the method buildRNN.

Slots

call

An object of class call returning an unevaluated function call.

model_path

The directory where the trained rnn model should be stored.

compile_options

A list contanning compile options of keras model, check buildRNN for more detail.

optim_hyperparameters

A list contanning tunable hyperparameters, check buildRNN.

preprocess_parameter

A character value indicating data preparation method, buildRNN.

tracking

A data frame containing the tracking of the gridsearch history of keras model.

Note

Check the Detail section of buildRNN to understand how SM.rnn model is created.

Functions

Available functions for this class are:

• show: display the information of the RNN model

• RNN_predict: predict function.

4301350/sm4sd documentation built on June 14, 2020, 4:22 p.m.