Skip to content
/ STRNN Public

Predicting the Next Location: A Recurrent Model with Spatial and Temporal Contexts

Notifications You must be signed in to change notification settings

yongqyu/STRNN

Repository files navigation

STRNN

Predicting the Next Location: A Recurrent Model with Spatial and Temporal Contexts
Qiang Liu, Shu Wu, Liang Wang, Tieniu Tan (AAAI-16)

Performance of this implementation

(recall@1: 0.0017160630653176505)
(recall@5: 0.003539380072217654)
(recall@10: 0.004004147152407851)
(recall@100: 0.00464767080190197)
(recall@1000: 0.01394045534150613)
(recall@10000: 0.07866900175131349)

Requirements

1.python 3.5
2.pytorch 0.4.0

Usage

0. Preprocessing

If you use (prepro_xxx_50.txt), you do not need to preprocess.
If you want to perform personally or modify the source, Gowalla is required at ../dataset/.

$ python preprocess.py

1. Training

$ python train_torch.py

About

Predicting the Next Location: A Recurrent Model with Spatial and Temporal Contexts

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages