I’ve been looking for a method to label data that is currently unlabeled based on portion of data that is already labeled, i’ve been researching, it has to do with semi-supervised learning. But the only example i found was using SGAN for image classification but all my data are real numbers.
For example my data consists of 6000 samples of 15 features for every sample and every feature corresponds to values between -100 and 1000. 4500 labels are positive class, 500 are negative class and 1000 are unlabeled. I would like to know a proper method for labeling the 1000 data.
Source: Python-3x Questions