cuTENSOR (nvmath. bindings. cutensor)#
For detailed documentation on the original C APIs, refer to the cuTENSOR documentation.
Enums and constants#
See |
|
|
|
|
This enum captures all unary and binary element-wise operations supported by the cuTENSOR library. |
|
cuTENSOR status type returnsThe type is used for function status returns. |
|
Allows users to specify the algorithm to be used for performing the desired tensor operation. |
|
This enum gives users finer control over the suggested workspace.This enum gives users finer control over the amount of workspace that is suggested by |
|
This enum lists all attributes of a |
|
This enum lists all attributes of a |
|
This enum determines the mode w.r.t. |
|
This enum determines the mode w.r.t. |
|
This enum defines what is considered a cache hit. |
|
This enum lists all attributes of a |
Functions#
|
Initializes the cuTENSOR library and allocates the memory for the library context. |
|
Frees all resources related to the provided library handle. |
|
Resizes the plan cache. |
Writes the Plan-Cache (that belongs to the provided handle) to file. |
|
Reads a Plan-Cache from file and overwrites the cachelines of the provided handle. |
|
|
Writes the --per library-- kernel cache to file. |
|
Reads a kernel cache from file and adds all non-existing JIT compiled kernels to the kernel cache. |
|
Creates a tensor descriptor. |
|
Frees all resources related to the provided tensor descriptor. |
|
This function creates an operation descriptor that encodes an elementwise trinary operation. |
|
Performs an element-wise tensor operation for three input tensors (see |
|
This function creates an operation descriptor for an elementwise binary operation. |
|
Performs an element-wise tensor operation for two input tensors (see |
|
This function creates an operation descriptor for a tensor permutation. |
|
Performs the tensor permutation that is encoded by |
|
This function allocates a cutensorOperationDescriptor_t object that encodes a tensor contraction of the form . |
|
Frees all resources related to the provided descriptor. |
Get the Python data type of the corresponding OperationDescriptorAttribute attribute. |
|
Set attribute of a cutensorOperationDescriptor_t object. |
|
This function retrieves an attribute of the provided cutensorOperationDescriptor_t object (see |
|
|
Allocates the cutensorPlanPreference_t, enabling users to limit the applicable kernels for a given plan/operation. |
|
Frees all resources related to the provided preference. |
|
Get the Python data type of the corresponding PlanPreferenceAttribute attribute. |
Set attribute of a cutensorPlanPreference_t object. |
|
Get attribute of a cutensorPlanPreference_t object. |
|
|
Get the Python data type of the corresponding PlanAttribute attribute. |
|
Retrieves information about an already-created plan (see |
|
Determines the required workspaceSize for the given operation encoded by |
|
This function allocates a cutensorPlan_t object, selects an appropriate kernel for a given operation (encoded by |
|
Frees all resources related to the provided plan. |
|
This routine computes the tensor contraction . |
|
Creates a cutensorOperatorDescriptor_t object that encodes a tensor reduction of the form . |
|
Performs the tensor reduction that is encoded by |
|
This function allocates a cutensorOperationDescriptor_t object that encodes a tensor contraction of the form . |
|
This routine computes the tensor contraction . |
Create a block-sparse tensor descriptor. |
|
Frees all resources related to the provided block-sparse tensor descriptor. |
|
This function allocates a cutensorOperationDescriptor_t object that encodes a block-sparse tensor contraction of the form . |
|
|
This routine computes the block-sparse tensor contraction . |
|
Returns the description string for an error code. |
Returns Version number of the CUTENSOR library. |
|
Returns version number of the CUDA runtime that cuTENSOR was compiled against. |
|
|
This function sets the logging output file. |
|
This function opens a logging output file in the given path. |
|
This function sets the value of the logging level. |
|
This function sets the value of the log mask. |
This function disables logging for the entire run. |