AutoNMT: A Framework to Streamline the Research of Seq2Seq Models

TL;DR

AutoNMT is a comprehensive framework that automates data handling, experimentation, and reporting for seq2seq models, facilitating research and customization in the field.

Contribution

It introduces an automated, toolkit-agnostic framework that simplifies and accelerates seq2seq model research and development.

Findings

Streamlines data preprocessing and analysis

Supports multiple seq2seq toolkits including Fairseq and OpenNMT

Provides automated report generation

Abstract

We present AutoNMT, a framework to streamline the research of seq-to-seq models by automating the data pipeline (i.e., file management, data preprocessing, and exploratory analysis), automating experimentation in a toolkit-agnostic manner, which allows users to use either their own models or existing seq-to-seq toolkits such as Fairseq or OpenNMT, and finally, automating the report generation (plots and summaries). Furthermore, this library comes with its own seq-to-seq toolkit so that users can easily customize it for non-standard tasks.