Files
armadillo-code/include/armadillo_bits/op_diagmat_meat.hpp
T
2021-09-21 14:18:36 +10:00

768 lines
20 KiB
C++

// SPDX-License-Identifier: Apache-2.0
//
// Copyright 2008-2016 Conrad Sanderson (http://conradsanderson.id.au)
// Copyright 2008-2016 National ICT Australia (NICTA)
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
// ------------------------------------------------------------------------
//! \addtogroup op_diagmat
//! @{
template<typename T1>
inline
void
op_diagmat::apply(Mat<typename T1::elem_type>& out, const Op<T1, op_diagmat>& X)
{
arma_extra_debug_sigprint();
typedef typename T1::elem_type eT;
if(is_Mat<T1>::value)
{
// allow detection of in-place operation
const unwrap<T1> U(X.m);
const Mat<eT>& A = U.M;
if(&out != &A) // no aliasing
{
const Proxy< Mat<eT> > P(A);
op_diagmat::apply(out, P);
}
else // we have aliasing
{
const uword n_rows = out.n_rows;
const uword n_cols = out.n_cols;
if((n_rows == 1) || (n_cols == 1)) // create diagonal matrix from vector
{
const eT* out_mem = out.memptr();
const uword N = out.n_elem;
Mat<eT> tmp(N,N, arma_zeros_indicator());
for(uword i=0; i<N; ++i) { tmp.at(i,i) = out_mem[i]; }
out.steal_mem(tmp);
}
else // create diagonal matrix from matrix
{
const uword N = (std::min)(n_rows, n_cols);
for(uword i=0; i < n_cols; ++i)
{
if(i < N)
{
eT& out_ii = out.at(i,i);
const eT val = out_ii;
arrayops::fill_zeros(out.colptr(i), n_rows);
out_ii = val;
}
else
{
arrayops::fill_zeros(out.colptr(i), n_rows);
}
}
}
}
}
else
{
const Proxy<T1> P(X.m);
if(P.is_alias(out))
{
Mat<eT> tmp;
op_diagmat::apply(tmp, P);
out.steal_mem(tmp);
}
else
{
op_diagmat::apply(out, P);
}
}
}
template<typename T1>
inline
void
op_diagmat::apply(Mat<typename T1::elem_type>& out, const Proxy<T1>& P)
{
arma_extra_debug_sigprint();
const uword n_rows = P.get_n_rows();
const uword n_cols = P.get_n_cols();
const uword n_elem = P.get_n_elem();
if(n_elem == 0) { out.reset(); return; }
const bool P_is_vec = (T1::is_row) || (T1::is_col) || (n_rows == 1) || (n_cols == 1);
if(P_is_vec)
{
out.zeros(n_elem, n_elem);
if(Proxy<T1>::use_at == false)
{
typename Proxy<T1>::ea_type Pea = P.get_ea();
for(uword i=0; i < n_elem; ++i) { out.at(i,i) = Pea[i]; }
}
else
{
if(n_rows == 1)
{
for(uword i=0; i < n_elem; ++i) { out.at(i,i) = P.at(0,i); }
}
else
{
for(uword i=0; i < n_elem; ++i) { out.at(i,i) = P.at(i,0); }
}
}
}
else // P represents a matrix
{
out.zeros(n_rows, n_cols);
const uword N = (std::min)(n_rows, n_cols);
for(uword i=0; i<N; ++i) { out.at(i,i) = P.at(i,i); }
}
}
template<typename T1, typename T2>
inline
void
op_diagmat::apply(Mat<typename T1::elem_type>& out, const Op< Glue<T1,T2,glue_times>, op_diagmat>& X)
{
arma_extra_debug_sigprint();
op_diagmat::apply_times(out, X.m.A, X.m.B);
}
template<typename T1, typename T2>
inline
void
op_diagmat::apply_times(Mat<typename T1::elem_type>& actual_out, const T1& X, const T2& Y, const typename arma_not_cx<typename T1::elem_type>::result* junk)
{
arma_extra_debug_sigprint();
arma_ignore(junk);
typedef typename T1::elem_type eT;
const partial_unwrap<T1> UA(X);
const partial_unwrap<T2> UB(Y);
const typename partial_unwrap<T1>::stored_type& A = UA.M;
const typename partial_unwrap<T2>::stored_type& B = UB.M;
arma_debug_assert_trans_mul_size< partial_unwrap<T1>::do_trans, partial_unwrap<T2>::do_trans >(A.n_rows, A.n_cols, B.n_rows, B.n_cols, "matrix multiplication");
const bool use_alpha = partial_unwrap<T1>::do_times || partial_unwrap<T2>::do_times;
const eT alpha = use_alpha ? (UA.get_val() * UB.get_val()) : eT(0);
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;
// check if the multiplication results in a vector
if( (partial_unwrap<T1>::do_trans == false) && (partial_unwrap<T2>::do_trans == false) )
{
if((A_n_rows == 1) || (B_n_cols == 1))
{
arma_extra_debug_print("trans_A = false; trans_B = false; vector result");
const Mat<eT> C = A*B;
const eT* C_mem = C.memptr();
const uword N = C.n_elem;
actual_out.zeros(N,N);
for(uword i=0; i<N; ++i) { actual_out.at(i,i) = (use_alpha) ? eT(alpha * C_mem[i]) : eT(C_mem[i]); }
return;
}
}
else
if( (partial_unwrap<T1>::do_trans == true ) && (partial_unwrap<T2>::do_trans == false) )
{
if((A_n_cols == 1) || (B_n_cols == 1))
{
arma_extra_debug_print("trans_A = true; trans_B = false; vector result");
const Mat<eT> C = trans(A)*B;
const eT* C_mem = C.memptr();
const uword N = C.n_elem;
actual_out.zeros(N,N);
for(uword i=0; i<N; ++i) { actual_out.at(i,i) = (use_alpha) ? eT(alpha * C_mem[i]) : eT(C_mem[i]); }
return;
}
}
else
if( (partial_unwrap<T1>::do_trans == false) && (partial_unwrap<T2>::do_trans == true ) )
{
if((A_n_rows == 1) || (B_n_rows == 1))
{
arma_extra_debug_print("trans_A = false; trans_B = true; vector result");
const Mat<eT> C = A*trans(B);
const eT* C_mem = C.memptr();
const uword N = C.n_elem;
actual_out.zeros(N,N);
for(uword i=0; i<N; ++i) { actual_out.at(i,i) = (use_alpha) ? eT(alpha * C_mem[i]) : eT(C_mem[i]); }
return;
}
}
else
if( (partial_unwrap<T1>::do_trans == true ) && (partial_unwrap<T2>::do_trans == true ) )
{
if((A_n_cols == 1) || (B_n_rows == 1))
{
arma_extra_debug_print("trans_A = true; trans_B = true; vector result");
const Mat<eT> C = trans(A)*trans(B);
const eT* C_mem = C.memptr();
const uword N = C.n_elem;
actual_out.zeros(N,N);
for(uword i=0; i<N; ++i) { actual_out.at(i,i) = (use_alpha) ? eT(alpha * C_mem[i]) : eT(C_mem[i]); }
return;
}
}
// if we got to this point, the multiplication results in a matrix
const bool is_alias = (UA.is_alias(actual_out) || UB.is_alias(actual_out));
Mat<eT> tmp;
Mat<eT>& out = (is_alias) ? tmp : actual_out;
if( (partial_unwrap<T1>::do_trans == false) && (partial_unwrap<T2>::do_trans == false) )
{
arma_extra_debug_print("trans_A = false; trans_B = false; matrix result");
out.zeros(A_n_rows, B_n_cols);
const uword N = (std::min)(A_n_rows, B_n_cols);
for(uword k=0; k < N; ++k)
{
eT acc1 = eT(0);
eT acc2 = eT(0);
const eT* B_colptr = B.colptr(k);
// condition: A_n_cols = B_n_rows
uword j;
for(j=1; j < A_n_cols; j+=2)
{
const uword i = (j-1);
const eT tmp_i = B_colptr[i];
const eT tmp_j = B_colptr[j];
acc1 += A.at(k, i) * tmp_i;
acc2 += A.at(k, j) * tmp_j;
}
const uword i = (j-1);
if(i < A_n_cols)
{
acc1 += A.at(k, i) * B_colptr[i];
}
const eT acc = acc1 + acc2;
out.at(k,k) = (use_alpha) ? eT(alpha * acc) : eT(acc);
}
}
else
if( (partial_unwrap<T1>::do_trans == true ) && (partial_unwrap<T2>::do_trans == false) )
{
arma_extra_debug_print("trans_A = true; trans_B = false; matrix result");
out.zeros(A_n_cols, B_n_cols);
const uword N = (std::min)(A_n_cols, B_n_cols);
for(uword k=0; k < N; ++k)
{
const eT* A_colptr = A.colptr(k);
const eT* B_colptr = B.colptr(k);
// condition: A_n_rows = B_n_rows
const eT acc = op_dot::direct_dot(A_n_rows, A_colptr, B_colptr);
out.at(k,k) = (use_alpha) ? eT(alpha * acc) : eT(acc);
}
}
else
if( (partial_unwrap<T1>::do_trans == false) && (partial_unwrap<T2>::do_trans == true ) )
{
arma_extra_debug_print("trans_A = false; trans_B = true; matrix result");
out.zeros(A_n_rows, B_n_rows);
const uword N = (std::min)(A_n_rows, B_n_rows);
for(uword k=0; k < N; ++k)
{
eT acc = eT(0);
// condition: A_n_cols = B_n_cols
for(uword i=0; i < A_n_cols; ++i)
{
acc += A.at(k,i) * B.at(k,i);
}
out.at(k,k) = (use_alpha) ? eT(alpha * acc) : eT(acc);
}
}
else
if( (partial_unwrap<T1>::do_trans == true ) && (partial_unwrap<T2>::do_trans == true ) )
{
arma_extra_debug_print("trans_A = true; trans_B = true; matrix result");
out.zeros(A_n_cols, B_n_rows);
const uword N = (std::min)(A_n_cols, B_n_rows);
for(uword k=0; k < N; ++k)
{
eT acc = eT(0);
const eT* A_colptr = A.colptr(k);
// condition: A_n_rows = B_n_cols
for(uword i=0; i < A_n_rows; ++i)
{
acc += A_colptr[i] * B.at(k,i);
}
out.at(k,k) = (use_alpha) ? eT(alpha * acc) : eT(acc);
}
}
if(is_alias) { actual_out.steal_mem(tmp); }
}
template<typename T1, typename T2>
inline
void
op_diagmat::apply_times(Mat<typename T1::elem_type>& actual_out, const T1& X, const T2& Y, const typename arma_cx_only<typename T1::elem_type>::result* junk)
{
arma_extra_debug_sigprint();
arma_ignore(junk);
typedef typename T1::pod_type T;
typedef typename T1::elem_type eT;
const partial_unwrap<T1> UA(X);
const partial_unwrap<T2> UB(Y);
const typename partial_unwrap<T1>::stored_type& A = UA.M;
const typename partial_unwrap<T2>::stored_type& B = UB.M;
arma_debug_assert_trans_mul_size< partial_unwrap<T1>::do_trans, partial_unwrap<T2>::do_trans >(A.n_rows, A.n_cols, B.n_rows, B.n_cols, "matrix multiplication");
const bool use_alpha = partial_unwrap<T1>::do_times || partial_unwrap<T2>::do_times;
const eT alpha = use_alpha ? (UA.get_val() * UB.get_val()) : eT(0);
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;
// check if the multiplication results in a vector
if( (partial_unwrap<T1>::do_trans == false) && (partial_unwrap<T2>::do_trans == false) )
{
if((A_n_rows == 1) || (B_n_cols == 1))
{
arma_extra_debug_print("trans_A = false; trans_B = false; vector result");
const Mat<eT> C = A*B;
const eT* C_mem = C.memptr();
const uword N = C.n_elem;
actual_out.zeros(N,N);
for(uword i=0; i<N; ++i) { actual_out.at(i,i) = (use_alpha) ? eT(alpha * C_mem[i]) : eT(C_mem[i]); }
return;
}
}
else
if( (partial_unwrap<T1>::do_trans == true ) && (partial_unwrap<T2>::do_trans == false) )
{
if((A_n_cols == 1) || (B_n_cols == 1))
{
arma_extra_debug_print("trans_A = true; trans_B = false; vector result");
const Mat<eT> C = trans(A)*B;
const eT* C_mem = C.memptr();
const uword N = C.n_elem;
actual_out.zeros(N,N);
for(uword i=0; i<N; ++i) { actual_out.at(i,i) = (use_alpha) ? eT(alpha * C_mem[i]) : eT(C_mem[i]); }
return;
}
}
else
if( (partial_unwrap<T1>::do_trans == false) && (partial_unwrap<T2>::do_trans == true ) )
{
if((A_n_rows == 1) || (B_n_rows == 1))
{
arma_extra_debug_print("trans_A = false; trans_B = true; vector result");
const Mat<eT> C = A*trans(B);
const eT* C_mem = C.memptr();
const uword N = C.n_elem;
actual_out.zeros(N,N);
for(uword i=0; i<N; ++i) { actual_out.at(i,i) = (use_alpha) ? eT(alpha * C_mem[i]) : eT(C_mem[i]); }
return;
}
}
else
if( (partial_unwrap<T1>::do_trans == true ) && (partial_unwrap<T2>::do_trans == true ) )
{
if((A_n_cols == 1) || (B_n_rows == 1))
{
arma_extra_debug_print("trans_A = true; trans_B = true; vector result");
const Mat<eT> C = trans(A)*trans(B);
const eT* C_mem = C.memptr();
const uword N = C.n_elem;
actual_out.zeros(N,N);
for(uword i=0; i<N; ++i) { actual_out.at(i,i) = (use_alpha) ? eT(alpha * C_mem[i]) : eT(C_mem[i]); }
return;
}
}
// if we got to this point, the multiplication results in a matrix
const bool is_alias = (UA.is_alias(actual_out) || UB.is_alias(actual_out));
Mat<eT> tmp;
Mat<eT>& out = (is_alias) ? tmp : actual_out;
if( (partial_unwrap<T1>::do_trans == false) && (partial_unwrap<T2>::do_trans == false) )
{
arma_extra_debug_print("trans_A = false; trans_B = false; matrix result");
out.zeros(A_n_rows, B_n_cols);
const uword N = (std::min)(A_n_rows, B_n_cols);
for(uword k=0; k < N; ++k)
{
T acc_real = T(0);
T acc_imag = T(0);
const eT* B_colptr = B.colptr(k);
// condition: A_n_cols = B_n_rows
for(uword i=0; i < A_n_cols; ++i)
{
// acc += A.at(k, i) * B_colptr[i];
const std::complex<T>& xx = A.at(k, i);
const std::complex<T>& yy = B_colptr[i];
const T a = xx.real();
const T b = xx.imag();
const T c = yy.real();
const T d = yy.imag();
acc_real += (a*c) - (b*d);
acc_imag += (a*d) + (b*c);
}
const eT acc = std::complex<T>(acc_real, acc_imag);
out.at(k,k) = (use_alpha) ? eT(alpha * acc) : eT(acc);
}
}
else
if( (partial_unwrap<T1>::do_trans == true) && (partial_unwrap<T2>::do_trans == false) )
{
arma_extra_debug_print("trans_A = true; trans_B = false; matrix result");
out.zeros(A_n_cols, B_n_cols);
const uword N = (std::min)(A_n_cols, B_n_cols);
for(uword k=0; k < N; ++k)
{
T acc_real = T(0);
T acc_imag = T(0);
const eT* A_colptr = A.colptr(k);
const eT* B_colptr = B.colptr(k);
// condition: A_n_rows = B_n_rows
for(uword i=0; i < A_n_rows; ++i)
{
// acc += std::conj(A_colptr[i]) * B_colptr[i];
const std::complex<T>& xx = A_colptr[i];
const std::complex<T>& yy = B_colptr[i];
const T a = xx.real();
const T b = xx.imag();
const T c = yy.real();
const T d = yy.imag();
// take into account the complex conjugate of xx
acc_real += (a*c) + (b*d);
acc_imag += (a*d) - (b*c);
}
const eT acc = std::complex<T>(acc_real, acc_imag);
out.at(k,k) = (use_alpha) ? eT(alpha * acc) : eT(acc);
}
}
else
if( (partial_unwrap<T1>::do_trans == false) && (partial_unwrap<T2>::do_trans == true) )
{
arma_extra_debug_print("trans_A = false; trans_B = true; matrix result");
out.zeros(A_n_rows, B_n_rows);
const uword N = (std::min)(A_n_rows, B_n_rows);
for(uword k=0; k < N; ++k)
{
T acc_real = T(0);
T acc_imag = T(0);
// condition: A_n_cols = B_n_cols
for(uword i=0; i < A_n_cols; ++i)
{
// acc += A.at(k,i) * std::conj(B.at(k,i));
const std::complex<T>& xx = A.at(k, i);
const std::complex<T>& yy = B.at(k, i);
const T a = xx.real();
const T b = xx.imag();
const T c = yy.real();
const T d = -yy.imag(); // take the conjugate
acc_real += (a*c) - (b*d);
acc_imag += (a*d) + (b*c);
}
const eT acc = std::complex<T>(acc_real, acc_imag);
out.at(k,k) = (use_alpha) ? eT(alpha * acc) : eT(acc);
}
}
else
if( (partial_unwrap<T1>::do_trans == true) && (partial_unwrap<T2>::do_trans == true) )
{
arma_extra_debug_print("trans_A = true; trans_B = true; matrix result");
out.zeros(A_n_cols, B_n_rows);
const uword N = (std::min)(A_n_cols, B_n_rows);
for(uword k=0; k < N; ++k)
{
T acc_real = T(0);
T acc_imag = T(0);
const eT* A_colptr = A.colptr(k);
// condition: A_n_rows = B_n_cols
for(uword i=0; i < A_n_rows; ++i)
{
// acc += std::conj(A_colptr[i]) * std::conj(B.at(k,i));
const std::complex<T>& xx = A_colptr[i];
const std::complex<T>& yy = B.at(k, i);
const T a = xx.real();
const T b = -xx.imag(); // take the conjugate
const T c = yy.real();
const T d = -yy.imag(); // take the conjugate
acc_real += (a*c) - (b*d);
acc_imag += (a*d) + (b*c);
}
const eT acc = std::complex<T>(acc_real, acc_imag);
out.at(k,k) = (use_alpha) ? eT(alpha * acc) : eT(acc);
}
}
if(is_alias) { actual_out.steal_mem(tmp); }
}
//
//
//
template<typename T1>
inline
void
op_diagmat2::apply(Mat<typename T1::elem_type>& out, const Op<T1, op_diagmat2>& X)
{
arma_extra_debug_sigprint();
typedef typename T1::elem_type eT;
const uword row_offset = X.aux_uword_a;
const uword col_offset = X.aux_uword_b;
const Proxy<T1> P(X.m);
if(P.is_alias(out))
{
Mat<eT> tmp;
op_diagmat2::apply(tmp, P, row_offset, col_offset);
out.steal_mem(tmp);
}
else
{
op_diagmat2::apply(out, P, row_offset, col_offset);
}
}
template<typename T1>
inline
void
op_diagmat2::apply(Mat<typename T1::elem_type>& out, const Proxy<T1>& P, const uword row_offset, const uword col_offset)
{
arma_extra_debug_sigprint();
const uword n_rows = P.get_n_rows();
const uword n_cols = P.get_n_cols();
const uword n_elem = P.get_n_elem();
if(n_elem == 0) { out.reset(); return; }
const bool P_is_vec = (T1::is_row) || (T1::is_col) || (n_rows == 1) || (n_cols == 1);
if(P_is_vec)
{
const uword n_pad = (std::max)(row_offset, col_offset);
out.zeros(n_elem + n_pad, n_elem + n_pad);
if(Proxy<T1>::use_at == false)
{
typename Proxy<T1>::ea_type Pea = P.get_ea();
for(uword i=0; i < n_elem; ++i) { out.at(row_offset + i, col_offset + i) = Pea[i]; }
}
else
{
if(n_rows == 1)
{
for(uword i=0; i < n_elem; ++i) { out.at(row_offset + i, col_offset + i) = P.at(0,i); }
}
else
{
for(uword i=0; i < n_elem; ++i) { out.at(row_offset + i, col_offset + i) = P.at(i,0); }
}
}
}
else // P represents a matrix
{
arma_debug_check_bounds
(
((row_offset > 0) && (row_offset >= n_rows)) || ((col_offset > 0) && (col_offset >= n_cols)),
"diagmat(): requested diagonal out of bounds"
);
out.zeros(n_rows, n_cols);
const uword N = (std::min)(n_rows - row_offset, n_cols - col_offset);
for(uword i=0; i<N; ++i)
{
const uword row = i + row_offset;
const uword col = i + col_offset;
out.at(row,col) = P.at(row,col);
}
}
}
//! @}