EXP: `finetuning_OOD_prediction`

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

EXP: `finetuning_OOD_prediction` #85

Closed MxMstrmn closed 2 years ago

MxMstrmn commented 2 years ago

Ref #67

This is WIP.

The lincs config add a lincs sweep on the 'split_ood_drugs' for all embeddings using the current best model hparams for each model type.

Additionally, I already added the config for the subsequent finetuning on the trapnell dataset. Here, the config.hash for each model has to be added. These should be analysed in analyse_all_embeddings_lincs_ood_drugs.ipynb once the first part of the experiment has been run successfully.

review-notebook-app[bot] commented 2 years ago

Check out this pull request on

See visual diffs & provide feedback on Jupyter Notebooks.

Powered by ReviewNB

MxMstrmn commented 2 years ago

I will close this as it is not relevant anymore.