First, please install the recent version of Pytorch and Torchvision as pip install torch torchvision. Then, you can install other package by running pip install -r requirements.txt
Please download our simulation data in the link_function and link_structure to reproduce the published results. Then, please decompress the floders as ./function_change/dataset/ and ./structure_change/dataset/ and replace the original floders as the downloaded ones.
The directory structure should look like
LLCP_simulation/
|–– function_change/
| |–– dataset
| |–– train.pkl
| |–– test.pkl
| |–– main.py
| |–– models_cvae.py
| |–– synthetic_simulate_data.py
| |–– utils.py
|–– structure_change/
| |–– dataset
| |–– train.pkl
| |–– test.pkl
| |–– main.py
| |–– models_cvae.py
| |–– synthetic_simulate_data.py
| |–– utils.py
If you would like to generate simulation data by yourself, please follow the below instructions.
For function change setting
cd function_change
python synthetic_simulate_data.py
For structure change setting
cd structure_change
python synthetic_simulate_data.py
Please run the following code to to reproduce the experiment results.
For function change setting
cd function_change
python main.py
For structure change setting
cd structure_change
python main.py