The UniNAS framework: combining modules in arbitrarily complex configurations with argument trees

TL;DR

UniNAS is a flexible framework for neural architecture search that uses argument trees for modular configuration, enabling complex, reusable setups and user-friendly experiment design without coding.

Contribution

The paper introduces argument trees as a novel modular configuration method, allowing flexible, complex, and reusable neural architecture search setups in UniNAS.

Findings

Supports complex module configurations with argument trees

Enables GUI-based experiment design without coding

Facilitates diverse NAS approaches within a unified framework

Abstract

Designing code to be simplistic yet to offer choice is a tightrope walk. Additional modules such as optimizers and data sets make a framework useful to a broader audience, but the added complexity quickly becomes a problem. Framework parameters may apply only to some modules but not others, be mutually exclusive or depend on each other, often in unclear ways. Even so, many frameworks are limited to a few specific use cases. This paper presents the underlying concept of UniNAS, a framework designed to incorporate a variety of Neural Architecture Search approaches. Since they differ in the number of optimizers and networks, hyper-parameter optimization, network designs, candidate operations, and more, a traditional approach can not solve the task. Instead, every module defines its own hyper-parameters and a local tree structure of module requirements. A configuration file specifies which modules are used, their used parameters, and which other modules they use in turn This concept of argument trees enables combining and reusing modules in complex configurations while avoiding many problems mentioned above. Argument trees can also be configured from a graphical user interface so that designing and changing experiments becomes possible without writing a single line of code. UniNAS is publicly available at https://github.com/cogsys-tuebingen/uninas