/LDNN

Theano implementation of Logistic Disjunctive Normal Networks.

Primary LanguagePython

LDNN

Theano implementation of Logistic Disjunctive Normal Networks.

Choose between two datasets of noisy_moons or noisy_circles by choosing data, labels = create_data(dataset='moons') or create_data(dataset='circles').

Choose between product approximation or min_max approximation of logical gates by choosing ldnn.classify(X, approx='product') or approx='min_max'

Reference

Sajjadi, Mehdi, Mojtaba Seyedhosseini, and Tolga Tasdizen. "Disjunctive Normal Networks." arXiv preprint arXiv:1412.8534 (2014).

http://arxiv.org/abs/1412.8534

Notice

This code is distributed without any warranty, express or implied.