Interface | Description |
---|---|
cublasLogCallback |
Interface for emulating a CUBLAS log callback.
|
Class | Description |
---|---|
cublasAtomicsMode |
The type indicates whether CUBLAS routines which has an alternate
implementation using atomics can be used.
|
cublasComputeType |
Enum for compute type
- default types provide best available performance using all available hardware features
and guarantee internal storage precision with at least the same precision and range;
- _PEDANTIC types ensure standard arithmetic and exact specified internal storage format;
- _FAST types allow for some loss of precision to enable higher throughput arithmetic.
|
cublasDataType | Deprecated
This class is replaced by
cudaDataType and only
kept for backward compatibility. |
cublasDiagType |
Indicates whether the main diagonal of the dense matrix is
unity and consequently should not be touched or modified
by the function.
|
cublasFillMode |
Indicates which part (lower or upper) of the dense matrix was
filled and consequently should be used by the function
|
cublasGemmAlgo |
Different GEMM algorithms
|
cublasHandle |
Java port of a cublasHandle
|
cublasMath |
Enum for default math mode/tensor operation
|
cublasOperation |
Indicates which operation needs to be performed with the
dense matrix.
|
cublasPointerMode |
Indicates whether the scalar values are passed by
reference on the host or device
|
cublasSideMode |
Indicates whether the dense matrix is on the left or right side
in the matrix equation solved by a particular function.
|
cublasStatus |
JCublas status return values.
|
JCublas |
Java bindings for CUBLAS, the NVIDIA CUDA BLAS library.
|
JCublas2 |
Java bindings for CUBLAS, the NVIDIA CUDA BLAS library.
|
This package contains classes that are related to JCublas.
The main class is the JCublas
class,
which provides all method bindings for CUBLAS, the
NVIDIA CUDA BLAS library.
Please consult your CUBLAS library documentation
for more information about how to use this API.
Copyright © 2020. All rights reserved.