LSTM using the iris dataset

Continuing with the LSTM architecture for RNN introduced in Chapter 6, Recurrent and Convolutional Neural Networks, we present the iris dataset processing using the mxnet LSTM function. The function expects all inputs and outputs as numeric. It is particularly useful for processing text sequences, but here we will train an LSTM model on the iris dataset. The input values are petal.length, petal.width, sepal.length, and sepal.width. The output variable is Species, which is converted to a numeric value between one and three. The iris dataset has been detailed in Chapter 4, Perceptron Neural Network Modeling – Basic Models:

#################################################################
### Chapter 7 - Neural Networks with R - Use cases      #########
### Prediction using LSTM on IRIS dataset               #########
#################################################################

##Required one time
library("mxnet")

data(iris)

x = iris[1:5!=5,-5]
y = as.integer...