FLA

This is our implementation for the paper:

Zhiyong Cheng, Fan Liu, Shenghan Mei, Yangyang Guo, Lei Zhu and Liqiang Nie. Feature-level Attentive ICF for Recommendation. Transactions on Information Systems (Under review).

Environment Settings

• Tensorflow-gpu version: 1.3.0

Example to run the codes.

Run FISM.py (delicious and Beauty)

python FISM.py --dataset delicious --batch_choice user --embed_size 16  --lr 0.01

python FISM.py --dataset Beauty --batch_choice user --embed_size 16  --lr 0.01

Run NAIS.py (delicious)

python NAIS.py --dataset delicious --batch_choice user --weight_size 16 --embed_size 16  --pretrain 1 --prelr 0.01 --lr 0.01 --beta 0.7

Run NAIS1.py (delicious)

python NAIS1.py --dataset delicious --batch_choice user --weight_size 16 --embed_size 16  --pretrain 1 --prelr 0.01 --lr 0.01 --beta 0.7

Run NAIS2.py (delicious)

python NAIS2.py --dataset delicious --batch_choice user --weight_size 16 --embed_size 16  --pretrain 1 --prelr 0.01 --lr 0.01 --beta 0.7