A look at sklearn’s LabelPropogation method to carry out semi-supervised learning

Python’s premier machine learning library, sklearn, so far has three functions to carry out semi-supervised learning, being the SelfTrainingClassifier, LabelPropogation, and LabelSpreading.

Semi-supervised learning is a situation where some of the labels in the training data are not labeled. Sklearn’s three semi-supervised estimators are able to make use of the unlabeled data to better capture the shape of the underlying data distribution and generalise better to new samples . These algorithms…