Neural Structured Learning (NSL) is a new open source tool to train neural networks by using Neural Graph Learning, utilizing structured data along with featured inputs. Structure can be explicit as represented by a graph or implicit as induced by adversarial perturbation.
NSL works with Tensorflow ML platform and can be easily used by both experienced and inexperience ML professional. NSL can train supervised, semi-supervised, or unsupervised learning for further model applications.
Read Full Article: https://www.tensorflow.org/neural_structured_learning