IBLBrainWideBenchMaskModelingSpikes#

class pretrain.datasets.IBLBrainWideBenchMaskModelingSpikes(root, split, dirname='ibl_brain_wide_bench_2026', recording_ids=None, transform=None)[source]#

Bases: core.dataset.IBLBrainWideBench2026

Binned spikes alone, for objectives that mask them and reconstruct them.

Samples the task-aligned windows of each pretrain recording, intersected with the split’s domain. No behavior is read and no target is returned: what to mask, and what to score it against, belong to the model.

Parameters:

  • root (str) – Directory holding the build.

  • split (Literal['train', 'val', 'test']) – Which pretrain split to read, train or val.

  • dirname (str) – Name of the build directory under root.

  • recording_ids (Union[str, list[str], None]) – Recordings to load, or None for every pretrain recording.

  • transform (Optional[Any]) – Applied to each item, the model’s input_fn among them.