---

Introduction

RecBole-MetaRec is an extended module for RecBole, which aims to help researches to compare and develop their own models in meta learning recommendation field.

This module is totally developed based on RecBole by adding extened classes and functions, without modifying any codes of RecBole core.

The contributions are briefly listed as follows:

• We extend MetaDataset from Dataset to split dataset by ‘task’.

• We extend MetaDataLoader from AbstractDataLoader to transform dataset into task form.

• We extend MetaRecommender from AbstractRecommender to provide a base recommender for implementing meta learning model.

• We extend MetaTrainer from Trainer to provide a base trainer for implementing meta learning training process.

• We extend MetaCollector from Collector to collect data for evaluation in meta learning circumstance.

• We implement MetaUtils with some useful toolkits for meta learning.

Therefore, researches can:

• Conveniently develop their own meta learning recommendation models.

• Conveniently learn and compare meta learning recommendation models that we have implemented.

• Enjoy advantages and features of RecBole.

Note: Before starting, it is strongly recommended to realize how RecBole works, and the homepage of RecBole is [https://recbole.io].

The construction is as following.

Get Started

• Installation
• Quick Start
• Versions

Models

• MeLU
• MAMO
• TaNP
• LWA
• NLBA
• MetaEmb
• MWUF

Develop Guide

• Overview
• Customize Configuration
• Customize Model
• Customize Trainer

Module Reference

• MetaDataset
• MetaDataLoader
• MetaRecommender
• MetaTrainer
• MetaCollector
• MetaUtils