TN Keras Layers

TN Keras exists to simplify tensorization of existing TensorFlow models. These layers try to match the APIs for existing Keras layers closely. Please note these layers are currently intended for experimentation only, not production. These layers are in alpha and upcoming releases might include breaking changes.

APIs are listed here. An overview of these layers is available here.

tensornetwork.tn_keras.layers.DenseDecomp(…) TN layer comparable to Dense that carries out matrix multiplication with 2 significantly smaller weight matrices instead of 1 large one.
tensornetwork.tn_keras.layers.DenseMPO(…) Matrix Product Operator (MPO) TN layer.
tensornetwork.tn_keras.layers.Conv2DMPO(…) 2D Convolutional Matrix Product Operator (MPO) TN layer.
tensornetwork.tn_keras.layers.DenseCondenser(…) Condenser TN layer.
tensornetwork.tn_keras.layers.DenseExpander(…) Expander TN layer.
tensornetwork.tn_keras.layers.DenseEntangler(…) Entangler TN layer.