BatchBench
A public benchmark and leaderboard for biomedical image analysis based on JUMP-CP Cell Painting data, employing a held-out batch protocol to evaluate model robustness to batch effects and domain shifts.
Timeline (last 90 days)
BatchBench is a public leaderboard based on the JUMP-CP dataset that specifically evaluates the robustness of biomedical image models on unseen experimental batches and data sources.
The core evaluation idea of BatchBench is to let participating models encounter a portion of experimental batches during training, and then test them on completely unseen batches and data sources (leave-batch-out design).
BatchBench provides a reproducible benchmark, giving research directions such as domain adaptation and robustness optimization a common reference framework.
The leave-batch-out setting of BatchBench is closer to domain generalization, i.e., the model never sees the target batch during training.
All Facts (4)
BatchBench is a public leaderboard based on the JUMP-CP dataset that specifically evaluates the robustness of biomedical image models on unseen experimental batches and data sources.
50%UnverifiedThe core evaluation idea of BatchBench is to let participating models encounter a portion of experimental batches during training, and then test them on completely unseen batches and data sources (leave-batch-out design).
50%UnverifiedBatchBench provides a reproducible benchmark, giving research directions such as domain adaptation and robustness optimization a common reference framework.
50%UnverifiedThe leave-batch-out setting of BatchBench is closer to domain generalization, i.e., the model never sees the target batch during training.
50%