This reposity contains all the codes to reproduce the results for our paper: Improved deep learning techniques in gravitational-wave data analysis by H Xia et al. (2020) [arxiv:2011.04418].
Our codes to generate gravitational-wave data are based on the open source tool ggwd for paper Convolutional neural networks: a magic bullet for gravitational-wave detection? [arXiv:1904.08693]. For details of using this tool, please check README.md in ./ggwd.
Environment needed to run our model including:
- PyTorch == 1.5.1
- numpy == 1.19.1
- scikit-learn == 0.23.2
- torchsummary == 1.5.1
- matplotlib == 3.3.1
- pandas == 1.1.3
- tqdm == 4.49.0
To settle the environment quickly, run:
pip install -r requirements.txt
Models that are used in our paper can be found in ./model. After generating your own datasets and preprocessing the data in ./ggwd, you can run run.ipynb to see the results of your own.