What is Multiclass Error Correcting Output Codes in shogun?

Recipe Objective

This recipe explains what is Multi-class Error-Correcting Output Codes explain with an example.
For more related projects-
Project 1
Project 2

Learn How to Build a Multi Class Text Classification Model using BERT

Multi-class Error-Correcting

ECOC (Error-Correcting Output Codes) is a multi-class learning strategy. ECOC trains 'L' binary classifier and transforms the results of the multiple classifications into a matrix, which is called the ECOC codebook.
We use a decoder to decode the codebook and to predict the labels of samples. We can encode or decode through multiple methods we have shown in the below code snippet.

x_train = RealFeatures(feats_train)
x_test = RealFeatures(feats_test)
y_train = MulticlassLabels(labels_train)
y_test = MulticlassLabels(labels_test)

c = LibLinear()
e = ECOCRandomDenseEncoder()
d = ECOCHDDecoder()
x = ECOCStrategy(e, d)
z = LinearMulticlassMachine(x, x_train, c, y_train)

z.train()
predict = z.apply_multiclass(x_test)
eval = MulticlassAccuracy()
accuracy = eval.evaluate(predict, y_test)