TensorKrowch: Smooth integration of tensor networks in machine learning

TL;DR

TensorKrowch is an open source Python library that simplifies the integration of tensor networks into machine learning models, enabling flexible construction, training, and deployment within deep learning pipelines.

Contribution

We introduce TensorKrowch, a user-friendly library that facilitates the incorporation of tensor networks into machine learning workflows with efficient implementation.

Findings

Enables construction and training of tensor networks in PyTorch

Provides seamless integration of tensor networks as deep learning layers

Optimized for efficient tensor network operations

Abstract

Tensor networks are factorizations of high-dimensional tensors into networks of smaller tensors. They have applications in physics and mathematics, and recently have been proposed as promising machine learning architectures. To ease the integration of tensor networks in machine learning pipelines, we introduce TensorKrowch, an open source Python library built on top of PyTorch. Providing a user-friendly interface, TensorKrowch allows users to construct any tensor network, train it, and integrate it as a layer in more intricate deep learning models. In this paper, we describe the main functionality and basic usage of TensorKrowch, and provide technical details on its building blocks and the optimizations performed to achieve efficient operation.