ClikaRT::ops::adaptive_avg_pool

function

adaptive_avg_pool()

Tensor adaptive_avg_pool(Tensor x, ClikaRT::Span<const std::int64_t> output_size)

Rank-generic adaptive average pooling to a target output size, channels-last.

The window geometry is DERIVED per output position so the spatial dims land exactly on output_size; no kernel/stride/padding to pick.

Parameters

• x: the input, [N, D1..Dn, C].
• output_size: target spatial extents, one per spatial dim.

Returns: [N, output_size.., C], dtype of x.

Throws

• ClikaRT::Error: (INVALID_ARGUMENT) when output_size's rank does not match the input's spatial rank.

Declared in ClikaRT/compute/ops.h, line 3822