Pseudo-label

Pseudo-labelingIt is the operation of adding prediction labels to unlabeled data when training the model. It solves the problem of unlabeled data and can also be used to train semi-supervised learning models together with labeled data.

The pseudo-labeling process diagram is as follows:

First, the model is trained on the labeled data, and then the trained model is used to predict the labels of the unlabeled data to create pseudo labels. After that, the labeled data and the newly generated pseudo-label data are integrated and used as new training data.