← Back to API reference

neuraltrain.models.common.SubjectLayers

pydantic model neuraltrain.models.common.SubjectLayers

Configuration for per-subject linear projections.

Parameters:

• n_subjects (int) – Number of subjects to allocate weight matrices for.

• bias (bool) – Include a bias term in each subject’s projection.

• init_id (bool) – Initialize projection matrices to the identity (requires in_channels == out_channels).

• mode ({"gather", "for_loop"}) – "gather" builds a (B, C_in, C_out) tensor via index_select (fast but memory-heavy for large channel counts). "for_loop" iterates over unique subjects (slower but lighter).

• subject_dropout (float or None) – Probability of replacing a subject index with a shared “dropout subject” during training. Required when average_subjects is True.

• average_subjects (bool) – At inference time, use the shared dropout-subject weights for all examples. Requires subject_dropout to be set.

Fields:

• average_subjects (bool)

• bias (bool)

• init_id (bool)

• mode (Literal['gather', 'for_loop'])

• n_subjects (int)

• subject_dropout (float | None)

field n_subjects: int = 200

field bias: bool = True

field init_id: bool = False

field mode: Literal['gather', 'for_loop'] = 'gather'

field subject_dropout: float | None = None

field average_subjects: bool = False

build(in_channels: int, out_channels: int) → Module

← Back to API reference