cd scripts/
python DataDownloader.py
python ModelTrainer.py allThen you can run each notebook and reproduce the results.
All datasets are available in this drive directory.
You can simply train each network with a specific dataset with the following scripts:
python -m scripts.train_trVAE kang[haber] python -m scripts.train_DCtrVAE mnist[celeba] python -m scripts.train_cvae kang[haber] python -m scripts.train_cyclegan kang[haber]python -m scripts.train_mdcvae kang[haber]python -m scripts.train_saucie kang[haber]python -m scripts.train_scGen kang[haber]python -m scripts.train_scVI kang[haber]| Study | notebook path |
|---|---|
| Haber et. al | Jupyter Notebooks/Haber.ipynb |
| Kang et. al | Jupyter Notebooks/Kang.ipynb |
| CelebA | Jupyter Notebooks/CelebA.ipynb |
| Figures | notebook path |
|---|---|
| Method Comparison - Haber et. al | Jupyter Notebooks/methodComparison-Haber.ipynb |
| Method Comparison - Kang et. al | Jupyter Notebooks/methodComparison-Kang.ipynb |
| Runtime Comparison - Kang et. al | Jupyter Notebooks/Time.ipynb |
| Simulation Response - Kang et. al | Jupyter Notebooks/BoxPlots_StackedViolins - Kang.ipynb |
To run the notebooks and scripts you need following packages :
tensorflow, scanpy, numpy, matplotlib, scipy, wget.