7.5: Deep Learning

Nowadays, “deep learning” is a bit of buzzword which used to designate software packages including multiple classification methods, and among the always some complicated neural networks (multi-layered, recurrent etc.) In that sense, R with necessary packages is a deep learning system. What is missed (actually, not), is a common interface to all “animals” in this zoo of methods. Package mlr was created to unify the learning interface in R:

Code \(\PageIndex{1}\) (R):

library(mlr)
...
## 1) Define the task
## Specify the type of analysis (e.g. classification)
## and provide data and response variable
task <- makeClassifTask(data=iris, target="Species")
## 2) Define the learner, use listLearners()[,1]
## Choose a specific algorithm
lrn <- makeLearner("classif.ctree")
n = nrow(iris)
train.set <- sample(n, size=2/3*n)
test.set <- setdiff(1:n, train.set)
## 3) Fit the model
## Train the learner on the task using a random subset
## of the data as training set
model <- train(lrn, task, subset=train.set)
## 4) Make predictions
## Predict values of the response variable for new
## observations by the trained model
## using the other part of the data as test set
pred <- predict(model, task=task, subset=test.set)
## 5) Evaluate the learner
## Calculate the mean misclassification error and accuracy
performance(pred, measures=list(mmce, acc))

In addition, R now has interfaces (ways to connect with) to (almost) all famous “deep learning” software systems, namely TensorFlow, H2O, Keras, Caffe and MXNet.