L_layer_model: Main function for multilayer NN estimation. Binary choice as...

In kostya75/mlbasic1: basic neural networks and logistic regression

Description

Main function for multilayer NN estimation. Binary choice as output layer. Hidden layers use 'relu', output layer is 'sigmoid'

Usage

L_layer_model(X, Y, layers_dims, learning_rate = 0.0075,
  num_iterations = 3000, print_cost = F)

Arguments

X	matrix of features. Each column represents one training example
Y	matrix (1 x m) of true responses
layers_dims	structure of the model
learning_rate	spcecify the learning rate for gradient descent
num_iterations	number of iterations for gradient descent
print_cost	boolean, whether or not to print cost

Examples

## Not run: 
first_layer<-nrow(X)
model_dim<-c(first_layer,,l2,l3,...,1)
parameters<-L_layer_model(X, Y, layers_dims=model_dim)

## End(Not run)

kostya75/mlbasic1 documentation built on June 21, 2020, 12:56 a.m.