The Compositional Perturbation Autoencoder (CPA) is a deep generative framework to learn effects of perturbations at the single-cell level. CPA performs OOD predictions of unseen combinations of drugs, learns interpretable embeddings, estimates dose-response curves, and provides uncertainty estimates.
Tuner added for hyperparameter optimization.
Example script and explanation available at examples/tune_script.py and README --> hyperparameter optimization
Tuner added for hyperparameter optimization. Example script and explanation available at
examples/tune_script.pyand README --> hyperparameter optimizationThe tuner is adapted and adjusted from scvi-tools v1.2.0: https://docs.scvi-tools.org/en/stable/release_notes/index.html
Fixes #45 Fixes #10 Fixes #7 Fixes #37