build_the_model: Build and compile keras sequential models with 2 hidden...

In RJ333/phyloseq2ML: Connects Phyloseq To Ranger And Keras

Description

This function can setup and compile sequential models. Please have a look at [keras_model_sequential](https://keras.rstudio.com/reference/keras_model_sequential.html), [layer_dense](https://keras.rstudio.com/reference/layer_dense.html) and [compile](https://keras.rstudio.com/reference/compile.html) for further details.

Usage

build_the_model(
  train_data,
  Layer1_units,
  Layer2_units,
  classes,
  Dropout_layer1,
  Dropout_layer2,
  Dense_activation_function,
  Output_activation_function = NULL,
  Optimizer_function,
  Loss_function,
  Metric,
  ...
)

Arguments

train_data	a table of training data
Layer1_units	integer, number of units in the first hidden layer
Layer2_units	integer, number of units in the second hidden layer
classes	integer, number of classes and therefore number of units in the output layer, set to 1 for regression
Dropout_layer1	numeric, ratio of dropout nodes for layer 1 between 0 and 1
Dropout_layer2	numeric, ratio of dropout nodes for layer 2 between 0 and 1
Dense_activation_function	char, activation function for the hidden layers
Output_activation_function	char, activation function for the output layer, (default: NULL, used for regression)
Optimizer_function	char, the optimizer function
Loss_function	char, the loss function
Metric	char vector, which metrics to monitor
...	further arguments

Value

a compiled keras sequential model with two hidden layers

RJ333/phyloseq2ML documentation built on June 2, 2020, 9:25 p.m.