fftconv1d_bhl_chunked

fftconv1d_bhl_chunked(x, kernel, shortcut=None, chunk_size=None)

Channel-chunked 1D non-causal convolution via subq_ops, BHL layout.

Parameters:

• x (Tensor) – Input tensor [B, H, L].

• kernel (Tensor) – Kernel tensor [1, H, K] or [H, K]; see fftconv1d_bhl().

• shortcut (Tensor | None) – Optional per-channel scale [H].

• chunk_size (int | None) – Channels per kernel call; None means 128.

Returns:

Output tensor [B, H, L] in x.dtype.

Return type:

Tensor