Mojo function

pack_conv_filter_shape

pack_conv_filter_shape[single_thread_blocking_override: Bool](filter: LayoutTensor[filter.dtype, filter.layout, filter.origin, address_space=filter.address_space, element_layout=filter.element_layout, layout_int_type=filter.layout_int_type, linear_idx_type=filter.linear_idx_type, masked=filter.masked, alignment=filter.alignment], num_groups: Int) -> IndexList[(filter.rank + 1)]

Compute the output shape of convolution filter packing.

Parameters:

single_thread_blocking_override (Bool): If True, then the operation is run synchronously using a single thread.

Args:

filter (LayoutTensor): The filter to be packed.
num_groups (Int): The number of groups in the convolution.

Returns:

IndexList: The output shape.