hidden_units=auto or list(integer)

Specifies the depth and number of interior nodes for the neural network underlying this classifier.

num_classes=auto or posint

Specifies the number of distinct categories into which the data should be classified.

The Classify command accepts a set of training data which has been classified into a finite set of classes, trains a neural network model for this classification, and returns a classifier function which can be applied to arbitrary additional data.

training_set := Import("example/iris_training.csv", base=datadir);

test_set := Import("example/iris_test.csv", base=datadir);

classifier := DeepLearning:-Classify( training_set[1..4], training_set[5] );

classifier( test_set[1..4], test_set[5] );

new_sample := DataSeries([4.9,3.1,1.5,0.1], labels=["SepalLength","SepalWidth","PetalLength","PetalWidth"]);

$\mathrm{new\_sample}≔\left[\begin{array}{cc}''SepalLength''& 4.9\\ ''SepalWidth''& 3.1\\ ''PetalLength''& 1.5\\ ''PetalWidth''& 0.1\end{array}\right]$

classifier( new_sample );

classifier( new_sample, output = probabilities );

The DeepLearning[Classify] command was introduced in Maple 2019.

For more information on Maple 2019 changes, see Updates in Maple 2019.