Evaluation: Add `evaluate_r2_sc` on per-cell level

theislab / chemCPA

Code for "Predicting Cellular Responses to Novel Drug Perturbations at a Single-Cell Resolution", NeurIPS 2022.

https://arxiv.org/abs/2204.13545

MIT License

88 stars 23 forks source link

Evaluation: Add `evaluate_r2_sc` on per-cell level #90

Closed siboehm closed 2 years ago

siboehm commented 2 years ago

If we check the perturbation disentanglement to make sure we're not accidentally training an autoencoder, then we can run a regular test-set reconstruction loss.

Otherwise we will not be able to discern models that perform well averaged-across-cells (eg by predicting the mean) from models that perform well on each individual cell.