AMOM

Introduction

This repo contains the code of AMOM, which can be applied to various sequence to sequence tasks.

Requirements and Installation

• PyTorch version == 1.10
• Python version >= 3.6
• Fairseq == 1.0.0
• sacrebleu==1.5.1
• sacremoses
• tensorboardX
• tensorboard

To install fairseq from source and develop locally:

cd fairseq-main
pip install --editable ./

Examples

• Nerual Translation Task.
• Summarization Task.
• Code Generation Task.