← Back to API reference

neuraltrain.models.simpleconv.SimpleConv

pydantic model neuraltrain.models.simpleconv.SimpleConv

1-D convolutional encoder, adapted from brainmagick.

Parameters:

• hidden (int) – Number of channels in the first convolutional layer.

• depth (int) – Number of convolutional layers.

• linear_out (bool) – Use a single transposed convolution as the output projection.

• complex_out (bool) – Use a two-layer transposed-convolution output projection with a non-linearity in between. Mutually exclusive with linear_out.

• kernel_size (int) – Kernel size for every convolutional layer (must be odd).

• growth (float) – Multiplicative channel growth factor per layer.

• dilation_growth (int) – Multiplicative dilation growth factor per layer.

• dilation_period (int or None) – If set, reset dilation to 1 every dilation_period layers.

• skip (bool) – Add residual skip connections when input and output shapes match.

• post_skip (bool) – Append a depth-wise convolution after each skip connection.

• scale (float or None) – If set, apply LayerScale with this initial value after each skip connection.

• rewrite (bool) – Append a 1x1 convolution + LeakyReLU after each layer.

• groups (int) – Number of groups for grouped convolutions (first layer always uses 1).

• glu (int) – If non-zero, insert a GLU gate every glu layers.

• glu_context (int) – Context (padding) size for the GLU convolution.

• glu_glu (bool) – If True the gate uses nn.GLU; otherwise the layer activation.

• gelu (bool) – Use GELU activation instead of (Leaky)ReLU.

• dropout (float) – Channel-dropout probability (currently raises NotImplementedError).

• dropout_rescale (bool) – Rescale activations after channel dropout.

• conv_dropout (float) – Dropout probability inside each convolutional block.

• dropout_input (float) – Dropout probability applied to the input of the convolutional stack.

• batch_norm (bool) – Apply batch normalization after each convolution.

• relu_leakiness (float) – Negative slope for LeakyReLU (0 gives standard ReLU).

• transformer_config (TransformerEncoder or None) – If set, append a Transformer encoder after the convolutional stack.

• subject_layers_config (SubjectLayers or None) – If set, prepend a per-subject linear projection.

• subject_layers_dim ({"input", "hidden"}) – Dimension used for the subject-layer projection.

• merger_config (ChannelMerger or None) – If set, prepend a ChannelMerger for multi-montage support.

• initial_linear (int) – If non-zero, prepend a 1x1 convolution projecting to this many channels.

• initial_depth (int) – Number of 1x1 convolution layers in the initial projection.

• initial_nonlin (bool) – Append a non-linearity after the initial 1x1 projection stack.

• backbone_out_channels (int or None) – If set, force the backbone output to this many channels.

Fields:

• backbone_out_channels (int | None)

• batch_norm (bool)

• complex_out (bool)

• conv_dropout (float)

• depth (int)

• dilation_growth (int)

• dilation_period (int | None)

• dropout (float)

• dropout_input (float)

• dropout_rescale (bool)

• gelu (bool)

• glu (int)

• glu_context (int)

• glu_glu (bool)

• groups (int)

• growth (float)

• hidden (int)

• initial_depth (int)

• initial_linear (int)

• initial_nonlin (bool)

• kernel_size (int)

• linear_out (bool)

• merger_config (neuraltrain.models.common.ChannelMerger | None)

• post_skip (bool)

• relu_leakiness (float)

• rewrite (bool)

• scale (float | None)

• skip (bool)

• subject_layers_config (neuraltrain.models.common.SubjectLayers | None)

• subject_layers_dim (Literal['input', 'hidden'])

• transformer_config (neuraltrain.models.transformer.TransformerEncoder | None)

field hidden: int = 16

field depth: int = 4

field linear_out: bool = False

field complex_out: bool = False

field kernel_size: int = 5

field growth: float = 1.0

field dilation_growth: int = 2

field dilation_period: int | None = None

field skip: bool = False

field post_skip: bool = False

field scale: float | None = None

field rewrite: bool = False

field groups: int = 1

field glu: int = 0

field glu_context: int = 0

field glu_glu: bool = True

field gelu: bool = False

field dropout: float = 0.0

field dropout_rescale: bool = True

field conv_dropout: float = 0.0

field dropout_input: float = 0.0

field batch_norm: bool = False

field relu_leakiness: float = 0.0

field transformer_config: TransformerEncoder | None = None

field subject_layers_config: SubjectLayers | None = None

field subject_layers_dim: Literal['input', 'hidden'] = 'hidden'

field merger_config: ChannelMerger | None = ChannelMerger(n_virtual_channels=270, fourier_emb_config=FourierEmb(n_freqs=None, total_dim=2048, n_dims=2, margin=0.2), dropout=0.2, dropout_around_channel=False, usage_penalty=0.0, n_subjects=200, per_subject=False, embed_ref=False, unmerge=False, invalid_value=-0.1)

field initial_linear: int = 0

field initial_depth: int = 1

field initial_nonlin: bool = False

field backbone_out_channels: int | None = None

build(n_in_channels: int, n_outputs: int) → SimpleConvModel

← Back to API reference