AutoRec Implementation

This repository contains a PyTorch implementation of the paper "AutoRec: Autoencoders Meet Collaborative Filtering". It is a model designed for collaborative filtering tasks in recommendation systems.

Requirements

Before running the code, ensure you have the required dependencies installed. You can install them using:

pip install -r requirements.txt

Running the Code

To train the AutoRec model, simply run the following command:

python main.py

Arguments

The main.py script accepts the following arguments:

Argument	Description	Default Value
-u	User-based training (default is item-based)	False
--data_folder	Path to the dataset folder	"ml-1m/"
--train_ratio	Ratio of the dataset to use for training	0.9
--early_stopping	Whether to use early stopping	True
--epochs	Number of training epochs	250
--batch_size	Batch size for training	512
--hsd	Hidden space dimension	500
--dropout_rate	Dropout rate for the AutoRec model	0.4
--lr	Learning rate for the optimizer	0.007
--weight_decay	Weight decay for the optimizer	8e-4

Dataset

The default dataset used is MovieLens 1M. Make sure the dataset is located in the folder specified by --data_folder. By default, this is ml-1m/.

If you want to use another dataset, place it in the desired folder and update the --data_folder argument.

Reference

If you use this implementation in your research, please cite the original paper:

1. Sedhain, S., Menon, A. K., Sanner, S., & Xie, L. (2015). AutoRec: Autoencoders Meet Collaborative Filtering. In Proceedings of the 24th International Conference on World Wide Web (pp. 111–112). Association for Computing Machinery. DOI: 10.1145/2740908.2742726