Modality-Aware Identity Construction and Counterfactual Structure Learning for ID-Free Multimodal Recommendation

Introduction

This is the Pytorch implementation for our paper: Modality-Aware Identity Construction and Counterfactual Structure Learning for ID-Free Multimodal Recommendation

Overview

The structure of our model is available for viewing in the following:

Enviroment Requirement

Python==3.8,

Pytorch==2.0.0,

Install all requirements with pip install -r requirements.txt.

Download data

Put the Baby, Clothing, Sports, Pet and Beauty datasets and other required data into the folder MAIL/data by downloading from this link Google Drive.

3. Training on a local server using PyCharm.

Run MAIL by python main.py with the default dataset as Baby. Specific dataset selection can be modified in main.py.

4. Modify specific parameters.

You may specify other parameters in CMD or config with configs/model/*.yaml and configs/dataset/*.yaml.

Acknowledgement

The structure of this code is based on MMRec and IDFREE. Thank for their work.