Model training method, device and equipment based on data augmentation

The embodiment of the invention discloses a model training method, device and equipment based on data augmentation. The method comprises the steps that a set of training samples is obtained, and the set comprises labeled samples and unlabeled samples; encoding to generate a first hidden variable corresponding to the labeled sample, and encoding to generate a second hidden variable corresponding to the unlabeled sample; generating a first classification result according to the first hidden variable, and determining a supervision loss value of the first classification result and the marked sample; decoding the second hidden variable to generate augmented data, and encoding the augmented data to generate a third hidden variable; generating a second classification result according to the second hidden variable, generating a third classification result according to the third hidden variable, and determining a consistency loss value of the second classification result and the third classification result; and fusing