← Back to API reference

neuraltrain.models.common.TemporalDownsampling

pydantic model neuraltrain.models.common.TemporalDownsampling

Temporal downsampling via a 2-D convolution over the time axis.

Parameters:

• kernel_size (int) – Kernel height (time-axis) of the downsampling convolution.

• stride (int) – Stride along the time axis.

• layer_norm (bool) – Apply LayerNorm after the convolution.

• layer_norm_affine (bool) – Use learnable affine parameters in the LayerNorm.

• gelu (bool) – Apply GELU activation after normalization.

Fields:

• gelu (bool)

• kernel_size (int)

• layer_norm (bool)

• layer_norm_affine (bool)

• stride (int)

field kernel_size: int = 45

field stride: int = 45

field layer_norm: bool = True

field layer_norm_affine: bool = True

field gelu: bool = True

build(dim: int) → Module

← Back to API reference