openvino.opset17#

Functions

bgr_to_nv12(arg[, single_plane, name])

Return a node which performs BGRtoNV12 operation.

erfinv(node[, name])

Return node which calculates the inverse error function element-wise on the input tensor.

grouped_matmul(mat_a, mat_b[, offsets, name])

Return a node which performs Grouped Matrix Multiplication for Mixture of Experts (MoE).

histc(data, *[, bins, min, max, ...])

Return a node which computes the histogram of all input elements using equal-width bins.

qsa_prepare(raw_index_keys, ...[, epsilon, name])

Return a node which builds one index key per complete key block for sparse attention.

rgb_to_nv12(arg[, single_plane, name])

Return a node which performs RGBtoNV12 operation.

sparse_attention(query, key, value, ...[, ...])

Return a node which computes attention over the key blocks selected for each query.

sparse_attention_indexer(index_query, ...[, ...])

Return a node which selects the key blocks every query token attends to.