cuquantum.cutensornet.gate_split¶
- cuquantum.cutensornet.gate_split(intptr_t handle, intptr_t desc_tensor_in_a, intptr_t raw_data_in_a, intptr_t desc_tensor_in_b, intptr_t raw_data_in_b, intptr_t desc_tensor_in_g, intptr_t raw_data_in_g, intptr_t desc_tensor_u, intptr_t u, intptr_t s, intptr_t desc_tensor_v, intptr_t v, int gate_algo, intptr_t svd_config, int compute_type, intptr_t svd_info, intptr_t work_desc, intptr_t stream)[source]¶
Performs gate split operation.
- Parameters
handle (intptr_t) – Opaque handle holding cuTensorNet’s library context.
desc_tensor_in_a (intptr_t) – Describes the modes, extents, and other metadata information of the input tensor A.
raw_data_in_a (intptr_t) – Pointer to the raw data of the input tensor A (in device memory).
desc_tensor_in_b (intptr_t) – Describes the modes, extents, and other metadata information of the input tensor B.
raw_data_in_b (intptr_t) – Pointer to the raw data of the input tensor B (in device memory).
desc_tensor_in_g (intptr_t) – Describes the modes, extents, and other metadata information of the input gate tensor.
raw_data_in_g (intptr_t) – Pointer to the raw data of the input gate tensor G (in device memory).
desc_tensor_u (intptr_t) – Describes the modes, extents, and other metadata information of the output U tensor. The extents of uncontracted modes are expected to be consistent with
desc_tensor_in_a
anddesc_tensor_in_g
.u (intptr_t) – Pointer to the output tensor data U (in device memory).
s (intptr_t) – Pointer to the output tensor data S (in device memory). Can be
NULL
when theCUTENSORNET_TENSOR_SVD_CONFIG_S_PARTITION
attribute ofsvd_config
is not set to default (::CUTENSORNET_TENSOR_SVD_PARTITION_NONE).desc_tensor_v (intptr_t) – Describes the modes, extents, and other metadata information of the output V tensor. The extents of uncontracted modes are expected to be consistent with
desc_tensor_in_b
anddesc_tensor_in_g
.v (intptr_t) – Pointer to the output tensor data V (in device memory).
gate_algo (int) – The algorithm to use for splitting the gate tensor into tensor A and B.
svd_config (intptr_t) – Opaque structure holding the user-requested SVD parameters.
compute_type (ComputeType) – Denotes the compute type used throughout the computation.
svd_info (intptr_t) – Opaque structure holding all information about the truncation at runtime.
work_desc (intptr_t) – Opaque structure describing the workspace.
stream (intptr_t) – The CUDA stream on which the computation is performed.
See also