TopoLSTM

Offer Pytorch(1.1.0) implementation

Our data preprocessing and topolstm structure refer to this repositary

How to Use ?

execute the following command to start training.

python train.py --data_dir '' --max_len 30 --batch_size 256

You can set parameter you want to tune your model(training)

--data_dir : your data directory
--max_len  : max diffusion path lengths
--batch_size : batch_size
--hidden_size : your embedding size
--epochs : training epochs
--learning_rate : optimizer learning rate

execute the following command to start testing.

python test.py --data_dir ''

You can set parameter you want to tune your model(testing)

--data_dir : your data directory

File Desription

preprocessing.py : preprocessing model need inputs
model.py : TopoLSTM implementation
train.py : Training TopoLSTM
test.py : Predict our test data

Use Data

You can get data from following link : data link

Citing

@inproceedings{WangZLC17,
  author    = {Jia Wang and
           Vincent W. Zheng and
           Zemin Liu and
           Kevin Chen-Chuan Chang},
  title     = {Topological Recurrent Neural Network for Diffusion Prediction},
  booktitle = {ICDM},
  pages     = {475--484},
  year      = {2017}
}