/// <summary>
/// Runs the m v float.
/// </summary>
/// <param name="result">The result.</param>
/// <param name="mat">The mat.</param>
/// <param name="vec">The vec.</param>
/// <exception cref="ArgumentException">lhs must be contiguous in the last dimension</exception>
private static void Run_M_V_float(Tensor result, Tensor mat, Tensor vec)
{
// Require lhs to be row-major. This means we must tell BLAS to transpose it (BLAS expects column-major matrices)
if (mat.Strides[1] != 1)
{
throw new ArgumentException("lhs must be contiguous in the last dimension");
}
unsafe
{
var yPtr = (float *)CpuNativeHelpers.GetBufferStart(result);
var aPtr = (float *)CpuNativeHelpers.GetBufferStart(mat);
var xPtr = (float *)CpuNativeHelpers.GetBufferStart(vec);
byte trans = (byte)'t';
int m = (int)mat.Shape[1];
int n = (int)mat.Shape[0];
int incx = (int)vec.Strides[0];
int lda = (int)mat.Strides[0];
int incy = (int)result.Strides[0];
float alpha = 1;
float beta = 0;
OpenBlasNative.sgemv_(&trans, &m, &n, &alpha, aPtr, &lda, xPtr, &incx, &beta, yPtr, &incy);
}
}
/// <summary>
/// Runs the m v double.
/// </summary>
/// <param name="result">The result.</param>
/// <param name="lhs">The LHS.</param>
/// <param name="rhs">The RHS.</param>
/// <exception cref="ArgumentException">lhs must be contiguous in the last dimension</exception>
private static void Run_M_V_double(Tensor result, Tensor lhs, Tensor rhs)
{
// Require lhs to be row-major. This means we must tell BLAS to transpose it (BLAS expects column-major matrices)
if (lhs.Strides[1] != 1)
{
throw new ArgumentException("lhs must be contiguous in the last dimension");
}
unsafe
{
var resultPtr = (double *)CpuNativeHelpers.GetBufferStart(result);
var lhsPtr = (double *)CpuNativeHelpers.GetBufferStart(lhs);
var rhsPtr = (double *)CpuNativeHelpers.GetBufferStart(rhs);
byte trans = (byte)'t';
int m = (int)rhs.Shape[1];
int n = (int)lhs.Shape[0];
int lda = (int)rhs.Strides[0];
int ldb = (int)lhs.Strides[0];
int ldc = (int)result.Strides[0];
double alpha = 1;
double beta = 0;
OpenBlasNative.dgemv_(&trans, &m, &n, &alpha, rhsPtr, &lda, lhsPtr, &ldb, &beta, resultPtr, &ldc);
}
}
private static void GemmOp(BlasOp transA, BlasOp transB, float alpha, Tensor a, Tensor b, float beta, Tensor c)
{
if (a.Strides[0] != 1)
{
throw new ArgumentException("a must be contiguous in the first dimension (column major / fortran order)");
}
if (b.Strides[0] != 1)
{
throw new ArgumentException("b must be contiguous in the first dimension (column major / fortran order)");
}
if (c.Strides[0] != 1)
{
throw new ArgumentException("c must be contiguous in the first dimension (column major / fortran order)");
}
unsafe
{
// dimensons: (m x k) * (k * n) = (m x n)
bool nta = transA == BlasOp.NonTranspose;
bool ntb = transB == BlasOp.NonTranspose;
byte transa = (byte)transA;
byte transb = (byte)transB;
int m = (int)a.Sizes[nta ? 0 : 1];
int k = (int)b.Sizes[ntb ? 0 : 1];
int n = (int)b.Sizes[ntb ? 1 : 0];
int lda = (int)a.Strides[1];
int ldb = (int)b.Strides[1];
int ldc = (int)c.Strides[1];
if (c.ElementType == DType.Float32)
{
float *aPtrSingle = (float *)CpuNativeHelpers.GetBufferStart(a);
float *bPtrSingle = (float *)CpuNativeHelpers.GetBufferStart(b);
float *cPtrSingle = (float *)CpuNativeHelpers.GetBufferStart(c);
SGEMM sgemm = new SGEMM();
sgemm.Run(System.Text.ASCIIEncoding.ASCII.GetString(&transa, 1), System.Text.ASCIIEncoding.ASCII.GetString(&transb, 1), m, n, k, alpha, aPtrSingle, lda, bPtrSingle, ldb, beta, cPtrSingle, ldc);
}
else if (c.ElementType == DType.Float64)
{
double *aPtrDouble = (double *)CpuNativeHelpers.GetBufferStart(a);
double *bPtrDouble = (double *)CpuNativeHelpers.GetBufferStart(b);
double *cPtrDouble = (double *)CpuNativeHelpers.GetBufferStart(c);
double alphaDouble = alpha;
double betaDouble = beta;
OpenBlasNative.dgemm_(&transa, &transb, &m, &n, &k, &alphaDouble, aPtrDouble, &lda, bPtrDouble, &ldb, &betaDouble, cPtrDouble, &ldc);
}
else
{
throw new NotSupportedException("CPU GEMM with element type " + c.ElementType + " not supported");
}
}
}
/// <summary>
/// Allocs the tensor reference.
/// </summary>
/// <param name="tensor">The tensor.</param>
/// <returns>TensorRef64.</returns>
public static TensorRef64 AllocTensorRef(NDArray tensor)
{
var tensorRef = new TensorRef64();
tensorRef.buffer = CpuNativeHelpers.GetBufferStart(tensor);
tensorRef.dimCount = tensor.Shape.Length;
tensorRef.sizes = AllocArray(tensor.Shape);
tensorRef.strides = AllocArray(tensor.Strides);
tensorRef.elementType = (CpuDType)tensor.ElementType;
return(tensorRef);
}
private static void GemmOp(BlasOp transA, BlasOp transB, float alpha, Tensor a, Tensor b, float beta, Tensor c)
{
if (a.Strides[0] != 1)
{
throw new ArgumentException("a must be contiguous in the first dimension (column major / fortran order)");
}
if (b.Strides[0] != 1)
{
throw new ArgumentException("b must be contiguous in the first dimension (column major / fortran order)");
}
if (c.Strides[0] != 1)
{
throw new ArgumentException("c must be contiguous in the first dimension (column major / fortran order)");
}
unsafe
{
// dimensons: (m x k) * (k * n) = (m x n)
var nta = transA == BlasOp.NonTranspose;
var ntb = transB == BlasOp.NonTranspose;
var transa = (byte)transA;
var transb = (byte)transB;
var m = (int)a.Sizes[nta ? 0 : 1];
var k = (int)b.Sizes[ntb ? 0 : 1];
var n = (int)b.Sizes[ntb ? 1 : 0];
var lda = (int)a.Strides[1];
var ldb = (int)b.Strides[1];
var ldc = (int)c.Strides[1];
if (c.ElementType == DType.Float32)
{
var aPtrSingle = (float *)CpuNativeHelpers.GetBufferStart(a);
var bPtrSingle = (float *)CpuNativeHelpers.GetBufferStart(b);
var cPtrSingle = (float *)CpuNativeHelpers.GetBufferStart(c);
OpenBlasNative.sgemm_(&transa, &transb, &m, &n, &k, &alpha, aPtrSingle, &lda, bPtrSingle, &ldb, &beta, cPtrSingle, &ldc);
}
else if (c.ElementType == DType.Float64)
{
var aPtrDouble = (double *)CpuNativeHelpers.GetBufferStart(a);
var bPtrDouble = (double *)CpuNativeHelpers.GetBufferStart(b);
var cPtrDouble = (double *)CpuNativeHelpers.GetBufferStart(c);
double alphaDouble = alpha;
double betaDouble = beta;
OpenBlasNative.dgemm_(&transa, &transb, &m, &n, &k, &alphaDouble, aPtrDouble, &lda, bPtrDouble, &ldb, &betaDouble, cPtrDouble, &ldc);
}
else
{
throw new NotSupportedException("CPU GEMM with element type " + c.ElementType + " not supported");
}
}
}
public static TensorRef64 AllocTensorRef(Tensor tensor)
{
TensorRef64 tensorRef = new TensorRef64
{
buffer = CpuNativeHelpers.GetBufferStart(tensor),
dimCount = tensor.Sizes.Length,
sizes = AllocArray(tensor.Sizes),
strides = AllocArray(tensor.Strides),
elementType = (CpuDType)tensor.ElementType
};
return(tensorRef);
}
/// <summary>
/// Runs the dot double.
/// </summary>
/// <param name="result">The result.</param>
/// <param name="lhs">The LHS.</param>
/// <param name="rhs">The RHS.</param>
private static void Run_Dot_double(Tensor result, Tensor lhs, Tensor rhs)
{
unsafe
{
var resultPtr = (double *)CpuNativeHelpers.GetBufferStart(result);
var lhsPtr = (double *)CpuNativeHelpers.GetBufferStart(lhs);
var rhsPtr = (double *)CpuNativeHelpers.GetBufferStart(rhs);
int n = (int)lhs.Shape[0];
int incx = (int)lhs.Strides[0];
int incy = (int)rhs.Strides[0];
* resultPtr = OpenBlasNative.ddot_(&n, lhsPtr, &incx, rhsPtr, &incy);
}
}
private static void Run_Dot_float(Tensor result, Tensor lhs, Tensor rhs)
{
unsafe
{
float *resultPtr = (float *)CpuNativeHelpers.GetBufferStart(result);
float *lhsPtr = (float *)CpuNativeHelpers.GetBufferStart(lhs);
float *rhsPtr = (float *)CpuNativeHelpers.GetBufferStart(rhs);
int n = (int)lhs.Sizes[0];
int incx = (int)lhs.Strides[0];
int incy = (int)rhs.Strides[0];
* resultPtr = OpenBlasNative.sdot_(&n, lhsPtr, &incx, rhsPtr, &incy);
}
}
/// <summary>
/// Runs the dot float.
/// </summary>
/// <param name="result">The result.</param>
/// <param name="lhs">The LHS.</param>
/// <param name="rhs">The RHS.</param>
private static void Run_Dot_float(NDArray result, NDArray lhs, NDArray rhs)
{
unsafe
{
var resultPtr = (float *)CpuNativeHelpers.GetBufferStart(result);
var lhsPtr = (float *)CpuNativeHelpers.GetBufferStart(lhs);
var rhsPtr = (float *)CpuNativeHelpers.GetBufferStart(rhs);
int n = (int)lhs.Shape[0];
int incx = (int)lhs.Strides[0];
int incy = (int)rhs.Strides[0];
* resultPtr = OpenBlasNative.sdot_(&n, lhsPtr, &incx, rhsPtr, &incy);
}
}
