neuraltrain.models.simpleconv.SimpleConvTimeAggModel¶

class neuraltrain.models.simpleconv.SimpleConvTimeAggModel(in_channels: int, out_channels: int, config: SimpleConvTimeAgg | None = None)[source][source]¶

nn.Module implementation of SimpleConvTimeAgg.

forward(x, subject_ids=None, channel_positions=None) → Tensor | dict[str, Tensor][source][source]¶

Run the convolutional encoder.

Parameters:

x (Tensor) – Input of shape (B, C, T).
subject_ids (Tensor or None) – Per-example subject indices, shape (B,).
channel_positions (Tensor or None) – Normalised electrode coordinates, shape (B, C, D).