341 lines
12 KiB
C++
341 lines
12 KiB
C++
// Copyright (C) 2008-2011 National ICT Australia (NICTA)
|
|
//
|
|
// This Source Code Form is subject to the terms of the Mozilla Public
|
|
// License, v. 2.0. If a copy of the MPL was not distributed with this
|
|
// file, You can obtain one at http://mozilla.org/MPL/2.0/.
|
|
// -------------------------------------------------------------------
|
|
//
|
|
// Written by Conrad Sanderson - http://conradsanderson.id.au
|
|
|
|
|
|
//! \addtogroup gemm_mixed
|
|
//! @{
|
|
|
|
|
|
|
|
//! \brief
|
|
//! Matrix multplication where the matrices have differing element types.
|
|
//! Uses caching for speedup.
|
|
//! Matrix 'C' is assumed to have been set to the correct size (i.e. taking into account transposes)
|
|
|
|
template<const bool do_trans_A=false, const bool do_trans_B=false, const bool use_alpha=false, const bool use_beta=false>
|
|
class gemm_mixed_large
|
|
{
|
|
public:
|
|
|
|
template<typename out_eT, typename in_eT1, typename in_eT2>
|
|
arma_hot
|
|
inline
|
|
static
|
|
void
|
|
apply
|
|
(
|
|
Mat<out_eT>& C,
|
|
const Mat<in_eT1>& A,
|
|
const Mat<in_eT2>& B,
|
|
const out_eT alpha = out_eT(1),
|
|
const out_eT beta = out_eT(0)
|
|
)
|
|
{
|
|
arma_extra_debug_sigprint();
|
|
|
|
const uword A_n_rows = A.n_rows;
|
|
const uword A_n_cols = A.n_cols;
|
|
|
|
const uword B_n_rows = B.n_rows;
|
|
const uword B_n_cols = B.n_cols;
|
|
|
|
if( (do_trans_A == false) && (do_trans_B == false) )
|
|
{
|
|
podarray<in_eT1> tmp(A_n_cols);
|
|
in_eT1* A_rowdata = tmp.memptr();
|
|
|
|
for(uword row_A=0; row_A < A_n_rows; ++row_A)
|
|
{
|
|
tmp.copy_row(A, row_A);
|
|
|
|
for(uword col_B=0; col_B < B_n_cols; ++col_B)
|
|
{
|
|
const in_eT2* B_coldata = B.colptr(col_B);
|
|
|
|
out_eT acc = out_eT(0);
|
|
for(uword i=0; i < B_n_rows; ++i)
|
|
{
|
|
acc += upgrade_val<in_eT1,in_eT2>::apply(A_rowdata[i]) * upgrade_val<in_eT1,in_eT2>::apply(B_coldata[i]);
|
|
}
|
|
|
|
if( (use_alpha == false) && (use_beta == false) ) { C.at(row_A,col_B) = acc; }
|
|
else if( (use_alpha == true ) && (use_beta == false) ) { C.at(row_A,col_B) = alpha*acc; }
|
|
else if( (use_alpha == false) && (use_beta == true ) ) { C.at(row_A,col_B) = acc + beta*C.at(row_A,col_B); }
|
|
else if( (use_alpha == true ) && (use_beta == true ) ) { C.at(row_A,col_B) = alpha*acc + beta*C.at(row_A,col_B); }
|
|
}
|
|
}
|
|
}
|
|
else
|
|
if( (do_trans_A == true) && (do_trans_B == false) )
|
|
{
|
|
for(uword col_A=0; col_A < A_n_cols; ++col_A)
|
|
{
|
|
// col_A is interpreted as row_A when storing the results in matrix C
|
|
|
|
const in_eT1* A_coldata = A.colptr(col_A);
|
|
|
|
for(uword col_B=0; col_B < B_n_cols; ++col_B)
|
|
{
|
|
const in_eT2* B_coldata = B.colptr(col_B);
|
|
|
|
out_eT acc = out_eT(0);
|
|
for(uword i=0; i < B_n_rows; ++i)
|
|
{
|
|
acc += upgrade_val<in_eT1,in_eT2>::apply(A_coldata[i]) * upgrade_val<in_eT1,in_eT2>::apply(B_coldata[i]);
|
|
}
|
|
|
|
if( (use_alpha == false) && (use_beta == false) ) { C.at(col_A,col_B) = acc; }
|
|
else if( (use_alpha == true ) && (use_beta == false) ) { C.at(col_A,col_B) = alpha*acc; }
|
|
else if( (use_alpha == false) && (use_beta == true ) ) { C.at(col_A,col_B) = acc + beta*C.at(col_A,col_B); }
|
|
else if( (use_alpha == true ) && (use_beta == true ) ) { C.at(col_A,col_B) = alpha*acc + beta*C.at(col_A,col_B); }
|
|
}
|
|
}
|
|
}
|
|
else
|
|
if( (do_trans_A == false) && (do_trans_B == true) )
|
|
{
|
|
Mat<in_eT2> B_tmp;
|
|
|
|
op_strans::apply_mat_noalias(B_tmp, B);
|
|
|
|
gemm_mixed_large<false, false, use_alpha, use_beta>::apply(C, A, B_tmp, alpha, beta);
|
|
}
|
|
else
|
|
if( (do_trans_A == true) && (do_trans_B == true) )
|
|
{
|
|
// mat B_tmp = trans(B);
|
|
// dgemm_arma<true, false, use_alpha, use_beta>::apply(C, A, B_tmp, alpha, beta);
|
|
|
|
|
|
// By using the trans(A)*trans(B) = trans(B*A) equivalency,
|
|
// transpose operations are not needed
|
|
|
|
podarray<in_eT2> tmp(B_n_cols);
|
|
in_eT2* B_rowdata = tmp.memptr();
|
|
|
|
for(uword row_B=0; row_B < B_n_rows; ++row_B)
|
|
{
|
|
tmp.copy_row(B, row_B);
|
|
|
|
for(uword col_A=0; col_A < A_n_cols; ++col_A)
|
|
{
|
|
const in_eT1* A_coldata = A.colptr(col_A);
|
|
|
|
out_eT acc = out_eT(0);
|
|
for(uword i=0; i < A_n_rows; ++i)
|
|
{
|
|
acc += upgrade_val<in_eT1,in_eT2>::apply(B_rowdata[i]) * upgrade_val<in_eT1,in_eT2>::apply(A_coldata[i]);
|
|
}
|
|
|
|
if( (use_alpha == false) && (use_beta == false) ) { C.at(col_A,row_B) = acc; }
|
|
else if( (use_alpha == true ) && (use_beta == false) ) { C.at(col_A,row_B) = alpha*acc; }
|
|
else if( (use_alpha == false) && (use_beta == true ) ) { C.at(col_A,row_B) = acc + beta*C.at(col_A,row_B); }
|
|
else if( (use_alpha == true ) && (use_beta == true ) ) { C.at(col_A,row_B) = alpha*acc + beta*C.at(col_A,row_B); }
|
|
}
|
|
}
|
|
|
|
}
|
|
}
|
|
|
|
};
|
|
|
|
|
|
|
|
//! Matrix multplication where the matrices have different element types.
|
|
//! Simple version (no caching).
|
|
//! Matrix 'C' is assumed to have been set to the correct size (i.e. taking into account transposes)
|
|
template<const bool do_trans_A=false, const bool do_trans_B=false, const bool use_alpha=false, const bool use_beta=false>
|
|
class gemm_mixed_small
|
|
{
|
|
public:
|
|
|
|
template<typename out_eT, typename in_eT1, typename in_eT2>
|
|
arma_hot
|
|
inline
|
|
static
|
|
void
|
|
apply
|
|
(
|
|
Mat<out_eT>& C,
|
|
const Mat<in_eT1>& A,
|
|
const Mat<in_eT2>& B,
|
|
const out_eT alpha = out_eT(1),
|
|
const out_eT beta = out_eT(0)
|
|
)
|
|
{
|
|
arma_extra_debug_sigprint();
|
|
|
|
const uword A_n_rows = A.n_rows;
|
|
const uword A_n_cols = A.n_cols;
|
|
|
|
const uword B_n_rows = B.n_rows;
|
|
const uword B_n_cols = B.n_cols;
|
|
|
|
if( (do_trans_A == false) && (do_trans_B == false) )
|
|
{
|
|
for(uword row_A = 0; row_A < A_n_rows; ++row_A)
|
|
{
|
|
for(uword col_B = 0; col_B < B_n_cols; ++col_B)
|
|
{
|
|
const in_eT2* B_coldata = B.colptr(col_B);
|
|
|
|
out_eT acc = out_eT(0);
|
|
for(uword i = 0; i < B_n_rows; ++i)
|
|
{
|
|
const out_eT val1 = upgrade_val<in_eT1,in_eT2>::apply(A.at(row_A,i));
|
|
const out_eT val2 = upgrade_val<in_eT1,in_eT2>::apply(B_coldata[i]);
|
|
acc += val1 * val2;
|
|
//acc += upgrade_val<in_eT1,in_eT2>::apply(A.at(row_A,i)) * upgrade_val<in_eT1,in_eT2>::apply(B_coldata[i]);
|
|
}
|
|
|
|
if( (use_alpha == false) && (use_beta == false) ) { C.at(row_A,col_B) = acc; }
|
|
else if( (use_alpha == true ) && (use_beta == false) ) { C.at(row_A,col_B) = alpha*acc; }
|
|
else if( (use_alpha == false) && (use_beta == true ) ) { C.at(row_A,col_B) = acc + beta*C.at(row_A,col_B); }
|
|
else if( (use_alpha == true ) && (use_beta == true ) ) { C.at(row_A,col_B) = alpha*acc + beta*C.at(row_A,col_B); }
|
|
}
|
|
}
|
|
}
|
|
else
|
|
if( (do_trans_A == true) && (do_trans_B == false) )
|
|
{
|
|
for(uword col_A=0; col_A < A_n_cols; ++col_A)
|
|
{
|
|
// col_A is interpreted as row_A when storing the results in matrix C
|
|
|
|
const in_eT1* A_coldata = A.colptr(col_A);
|
|
|
|
for(uword col_B=0; col_B < B_n_cols; ++col_B)
|
|
{
|
|
const in_eT2* B_coldata = B.colptr(col_B);
|
|
|
|
out_eT acc = out_eT(0);
|
|
for(uword i=0; i < B_n_rows; ++i)
|
|
{
|
|
acc += upgrade_val<in_eT1,in_eT2>::apply(A_coldata[i]) * upgrade_val<in_eT1,in_eT2>::apply(B_coldata[i]);
|
|
}
|
|
|
|
if( (use_alpha == false) && (use_beta == false) ) { C.at(col_A,col_B) = acc; }
|
|
else if( (use_alpha == true ) && (use_beta == false) ) { C.at(col_A,col_B) = alpha*acc; }
|
|
else if( (use_alpha == false) && (use_beta == true ) ) { C.at(col_A,col_B) = acc + beta*C.at(col_A,col_B); }
|
|
else if( (use_alpha == true ) && (use_beta == true ) ) { C.at(col_A,col_B) = alpha*acc + beta*C.at(col_A,col_B); }
|
|
}
|
|
}
|
|
}
|
|
else
|
|
if( (do_trans_A == false) && (do_trans_B == true) )
|
|
{
|
|
for(uword row_A = 0; row_A < A_n_rows; ++row_A)
|
|
{
|
|
for(uword row_B = 0; row_B < B_n_rows; ++row_B)
|
|
{
|
|
out_eT acc = out_eT(0);
|
|
for(uword i = 0; i < B_n_cols; ++i)
|
|
{
|
|
acc += upgrade_val<in_eT1,in_eT2>::apply(A.at(row_A,i)) * upgrade_val<in_eT1,in_eT2>::apply(B.at(row_B,i));
|
|
}
|
|
|
|
if( (use_alpha == false) && (use_beta == false) ) { C.at(row_A,row_B) = acc; }
|
|
else if( (use_alpha == true ) && (use_beta == false) ) { C.at(row_A,row_B) = alpha*acc; }
|
|
else if( (use_alpha == false) && (use_beta == true ) ) { C.at(row_A,row_B) = acc + beta*C.at(row_A,row_B); }
|
|
else if( (use_alpha == true ) && (use_beta == true ) ) { C.at(row_A,row_B) = alpha*acc + beta*C.at(row_A,row_B); }
|
|
}
|
|
}
|
|
}
|
|
else
|
|
if( (do_trans_A == true) && (do_trans_B == true) )
|
|
{
|
|
for(uword row_B=0; row_B < B_n_rows; ++row_B)
|
|
{
|
|
|
|
for(uword col_A=0; col_A < A_n_cols; ++col_A)
|
|
{
|
|
const in_eT1* A_coldata = A.colptr(col_A);
|
|
|
|
out_eT acc = out_eT(0);
|
|
for(uword i=0; i < A_n_rows; ++i)
|
|
{
|
|
acc += upgrade_val<in_eT1,in_eT2>::apply(B.at(row_B,i)) * upgrade_val<in_eT1,in_eT2>::apply(A_coldata[i]);
|
|
}
|
|
|
|
if( (use_alpha == false) && (use_beta == false) ) { C.at(col_A,row_B) = acc; }
|
|
else if( (use_alpha == true ) && (use_beta == false) ) { C.at(col_A,row_B) = alpha*acc; }
|
|
else if( (use_alpha == false) && (use_beta == true ) ) { C.at(col_A,row_B) = acc + beta*C.at(col_A,row_B); }
|
|
else if( (use_alpha == true ) && (use_beta == true ) ) { C.at(col_A,row_B) = alpha*acc + beta*C.at(col_A,row_B); }
|
|
}
|
|
}
|
|
|
|
}
|
|
}
|
|
|
|
};
|
|
|
|
|
|
|
|
|
|
|
|
//! \brief
|
|
//! Matrix multplication where the matrices have differing element types.
|
|
|
|
template<const bool do_trans_A=false, const bool do_trans_B=false, const bool use_alpha=false, const bool use_beta=false>
|
|
class gemm_mixed
|
|
{
|
|
public:
|
|
|
|
//! immediate multiplication of matrices A and B, storing the result in C
|
|
template<typename out_eT, typename in_eT1, typename in_eT2>
|
|
inline
|
|
static
|
|
void
|
|
apply
|
|
(
|
|
Mat<out_eT>& C,
|
|
const Mat<in_eT1>& A,
|
|
const Mat<in_eT2>& B,
|
|
const out_eT alpha = out_eT(1),
|
|
const out_eT beta = out_eT(0)
|
|
)
|
|
{
|
|
arma_extra_debug_sigprint();
|
|
|
|
Mat<in_eT1> tmp_A;
|
|
Mat<in_eT2> tmp_B;
|
|
|
|
const bool predo_trans_A = ( (do_trans_A == true) && (is_cx<in_eT1>::yes) );
|
|
const bool predo_trans_B = ( (do_trans_B == true) && (is_cx<in_eT2>::yes) );
|
|
|
|
if(do_trans_A)
|
|
{
|
|
op_htrans::apply_mat_noalias(tmp_A, A);
|
|
}
|
|
|
|
if(do_trans_B)
|
|
{
|
|
op_htrans::apply_mat_noalias(tmp_B, B);
|
|
}
|
|
|
|
const Mat<in_eT1>& AA = (predo_trans_A == false) ? A : tmp_A;
|
|
const Mat<in_eT2>& BB = (predo_trans_B == false) ? B : tmp_B;
|
|
|
|
if( (AA.n_elem <= 64u) && (BB.n_elem <= 64u) )
|
|
{
|
|
gemm_mixed_small<((predo_trans_A) ? false : do_trans_A), ((predo_trans_B) ? false : do_trans_B), use_alpha, use_beta>::apply(C, AA, BB, alpha, beta);
|
|
}
|
|
else
|
|
{
|
|
gemm_mixed_large<((predo_trans_A) ? false : do_trans_A), ((predo_trans_B) ? false : do_trans_B), use_alpha, use_beta>::apply(C, AA, BB, alpha, beta);
|
|
}
|
|
}
|
|
|
|
|
|
};
|
|
|
|
|
|
|
|
//! @}
|