Input Labels for Neural Networks - Deep Learning Dictionary
Labelling Datasets - Deep Learning Dictionary
During the supervised learning task of training a network to classify categorical data, we pass the samples and their corresponding labels to the network. The network uses the labels to calculate the loss between its predicted labels for the samples and the true labels.
When discussing labels, we typically have practical names for the labels, like dumbbells, kettlebells, barbell, etc. When training the network, however, these human-understandable word labels become encoded into numeric values. The network needs this encoding since mathematical operations will be occurring using these labels.
Committed by on