deep-learning-recipes

R implementation for selected machine learning methods with deep learning frameworks (Keras, Tensorflow).

Principles: clean, self-contained, minimal dependency, works with the latest framework versions.

Contents

• Triplet losses for implicit feedback recommender systems. [blog post] [code]
• Matrix factorization for binary implicit feedback data. [blog post] [code]
• "Wide and deep" model for regression/classification. [blog post] [code]