Deep Matrix Factorization Models for Recommender Systems

A Non-official Implementation of "Deep Matrix Factorization Models for Recommender Systems"

See paper: http://www.ijcai.org/proceedings/2017/0447.pdf

If you use the codes for your paper as baseline implementation, please cite the link: https://github.com/hegongshan/deep_matrix_factorization

Environment Settings

We use Keras with Tensorflow as the backend.

• Keras version: 2.3.0
• TensorFlow: 2.0.0

Example to run the codes.

python dmf.py --dataset ml-1m --user_layers [512,64] --item_layers [1024,64] --epochs 100 --lr 0.0001

Experimental Results

when epochs = 10 and lr = 0.001

	HR@10	NDCG@10	model file
ml-1m	0.5225	0.2930	model/ml-1m_u[512, 64]_i[1024, 64]_256_1572343913.h5

Tips: Each epoch takes about an hour and a half.

If you are interested in DMF, you can try to set lr to 0.0001 and run 100 epochs.

And then, HR@10 and NDCG@10 should be closer to the results in this paper.

Last Update: November 10, 2020