newswries-topics-classification

we’ll build a network to classify Reuters newswires into 46 mutually exclusive topics. Because we have many classes, this problem is an instance of multiclass classification; and because each data point should be classified into only one category, the problem is more specifically an instance of single-label, multiclass classification.