For the complete documentation index, see llms.txt. Markdown versions of all pages are available by appending .md to any URL (e.g. /max/get-started.md).

Mojo function

causal_conv1d_shape

def causal_conv1d_shape[dtype: DType, rank: Int](input: ManagedTensorSlice[IOSpec[_, _].Input, static_spec=input.static_spec], weight: ManagedTensorSlice[IOSpec[_, _].Input, static_spec=weight.static_spec], bias: ManagedTensorSlice[IOSpec[_, _].Input, static_spec=bias.static_spec]) -> IndexList[rank]

Returns the output shape for the causal_conv1d op.

Causal 1D convolution preserves the input shape: output has the same (batch, channels, seqlen) as input.

Parameters:

dtype (DType): Element type of the input, weight, and bias tensors.
rank (Int): Tensor rank of the input and output, expected to be 3.

Args:

input (ManagedTensorSlice[IOSpec[_, _].Input, static_spec=input.static_spec]): Input tensor with shape (batch, channels, seqlen).
weight (ManagedTensorSlice[IOSpec[_, _].Input, static_spec=weight.static_spec]): Convolution weights with shape (channels, width).
bias (ManagedTensorSlice[IOSpec[_, _].Input, static_spec=bias.static_spec]): Per-channel bias with shape (channels,).

Returns:

IndexList[rank]: The output tensor shape, equal to input.shape().