netket.nn.MaskedConv1D

class netket.nn.MaskedConv1D

Bases: Module

1D convolution module with mask for autoregressive NN.

Attributes

feature_group_count: int = 1

1).

Type:

if specified, divides the input features into groups (default

precision: Any = None

numerical precision of the computation, see jax.lax.Precision for details.

use_bias: bool = True

True).

Type:

whether to add a bias to the output (default

features: int

number of convolution filters.

kernel_size: int

length of the convolutional kernel.

kernel_dilation: int

dilation factor of the convolution kernel.

exclusive: bool

True if an output element does not depend on the input element at the same index.

kernel_init: Callable[[Any, Sequence[int], None | str | type[Any] | dtype | _SupportsDType], Array]

initializer for the convolutional kernel.

bias_init: Callable[[Any, Sequence[int], None | str | type[Any] | dtype | _SupportsDType], Array]

initializer for the bias.

Methods

__call__(inputs)

Applies a masked convolution to the inputs. For 1D convolution, there is not really a mask. We only need to apply appropriate padding.

Parameters:

inputs (Union[ndarray, Array]) – input data with dimensions (batch, length, features).

Return type:

Union[ndarray, Array]

Returns:

The convolved data.