Hands-On Deep Learning with R

Getting started with MXNet

For MXNet, we will use the same prepared adult census dataset. We will also use a multilayer perceptron as our model. Fitting a model using MXNet will be quite familiar to you if you are familiar with fitting a model using a number of other common machine learning packages:

1. First, we will load the MXNet package with the following line of code:

library(mxnet)

2. Then, we will define our multilayer perceptron. A seed value is set for reproducibility purposes. Afterward, the training data is converted to a data matrix and passed as an argument to the model, along with the training target values, as shown:

mx.set.seed(0)model <- mx.mlp(data.matrix(train), train_target, hidden_node=10, out_node=2, out_activation="softmax", ...