Neural Network Distiller: A Python Package For DNN Compression Research

TL;DR

Neural Network Distiller is an open-source Python library that provides a comprehensive suite of DNN compression algorithms, tools, and tutorials to support research and engineering in model compression.

Contribution

It introduces a modular, extensible Python package for DNN compression research, combining algorithms, tools, and tutorials in one accessible library.

Findings

Provides a versatile set of DNN compression algorithms.

Facilitates research with tutorials and sample applications.

Designed for extensibility and ease of use.

Abstract

This paper presents the philosophy, design and feature-set of Neural Network Distiller, an open-source Python package for DNN compression research. Distiller is a library of DNN compression algorithms implementations, with tools, tutorials and sample applications for various learning tasks. Its target users are both engineers and researchers, and the rich content is complemented by a design-for-extensibility to facilitate new research. Distiller is open-source and is available on Github at https://github.com/NervanaSystems/distiller.