The source code for the StageNet Reproducibility Project.

• StageNet: https://github.com/v1xerunt/StageNet
• T-LSTM: https://github.com/duskybomb/tlstm
• ON-LSTM: https://github.com/yikangshen/Ordered-Neurons

• Install Python, PyTorch. I used Python 3.8.8 and PyTorch 1.7.1.
• Install CUDA if you plan to use GPU acceleration.

(Adapted from original code repo)

MIMIC-III data is used for reproducing the StageNet and other models for this project. The raw data can be acquired through PhysioNet: https://physionet.org/content/mimiciii/1.4/. The raw data is not provided in this repo. You should first generate the decompensation benchmark dataset according to MIMIC-III Benchmarks: https://github.com/YerevaNN/mimic3-benchmarks/.

After building the decompensation benchmarks dataset, please save the files (.csv files and listfiles) in decompensation directory to data/ directory. Some sample pre-processed data files are saved in this folder for refernece purposes.

Note: You may need to run through the ValGenerator.ipynb file provided in the data folder to generate the validation dataset required by the StageNet model.

Trained models are saved in ./saved_weights

• StageNet model:

  $ python train.py --test_mode=1 --data_path='./data/'

• StageNet model with custom hyper-parameters (e.g. changing hyper-parameter K to 5):

  $ python train_StageNet_K5.py --test_mode=1 --data_path='./data/'

• LSTM model:

  $ python train_LSTM.py --test_mode=1 --data_path='./data/'

• T-LSTM model:

  $ python train_T_LSTM.py --test_mode=1 --data_path='./data/'

• StageNet-reduced model:

  $ python train_StageNet_I.py --test_mode=1 --data_path='./data/'

• StageNet model:

  $ python train.py --data_path='./data/ --file_name='trained_model'

• StageNet model with custom hyper-parameters (e.g. changing hyper-parameter K to 5):

  $ python train.py --data_path='./data/' --file_name='trained_model' --K=5

• LSTM model:

  $ python train_LSTM.py --data_path='./data/' --file_name='trained_model'

• T-LSTM model:

  $ python train_T_LSTM.py --data_path='./data/' --file_name='trained_model'

• StageNet-reduced model:

  $ python train_StageNet_I.py --data_path='./data/' --file_name='trained_model'

Junyi Gao, Cao Xiao, Yasha Wang, Wen Tang, Lucas M. Glass, and Jimeng Sun. 2020. StageNet: Stage-Aware Neural Networks for Health Risk Prediction. Proceedings of The Web Conference 2020. Association for Computing Machinery, New York, NY, USA, 530–540. DOI: https://doi.org/10.1145/3366423.3380136