void
check(vsip::const_Vector<T, BlockT> vec, int k, int shift=0)
{
using vsip::no_subblock;
using vsip::index_type;
typedef T value_type;
#if VERBOSE
std::cout << "check(k=" << k << ", shift=" << shift << "):"
<< std::endl;
#endif
if (subblock(vec) != no_subblock)
{
for (index_type li=0; li<vec.local().size(); ++li)
{
index_type gi = global_from_local_index(vec, 0, li);
#if VERBOSE
std::cout << " - " << li << " gi:" << gi << " = "
<< vec.local().get(li)
<< " exp: " << value<T>(gi + shift, k)
<< std::endl;
#endif
#if DO_ASSERT
test_assert(vec.local().get(li) == value<T>(gi + shift, k));
#endif
}
}
}
void
check_col_vector(vsip::const_Vector<T, BlockT> view, int col, int k=0)
{
using vsip::no_subblock;
using vsip::index_type;
typedef T value_type;
if (subblock(view) != no_subblock)
{
for (index_type lr=0; lr<view.local().size(0); ++lr)
{
index_type gr = global_from_local_index(view, 0, lr);
test_assert(view.local().get(lr) == value<T>(gr, col, k));
}
}
}
void
check_row_vector(vsip::const_Vector<T, BlockT> view, int row, int k=1)
{
using vsip::no_subblock;
using vsip::index_type;
typedef T value_type;
if (subblock(view) != no_subblock)
{
for (index_type lc=0; lc<view.local().size(0); ++lc)
{
index_type gc = global_from_local_index(view, 0, lc);
test_assert(view.local().get(lc) ==
value<T>(row, gc, k));
}
}
}
void
conv(
vsip::symmetry_type sym,
vsip::support_region_type sup,
vsip::const_Vector<T, Block1> coeff,
vsip::const_Vector<T, Block2> in,
vsip::Vector<T, Block3> out,
vsip::length_type D)
{
using vsip::index_type;
using vsip::length_type;
using vsip::stride_type;
using vsip::Vector;
using vsip::const_Vector;
using vsip::Domain;
using vsip::unbiased;
using vsip::impl::convert_to_local;
using vsip::impl::Working_view_holder;
typedef typename vsip::impl::scalar_of<T>::type scalar_type;
Working_view_holder<const_Vector<T, Block1> > w_coeff(coeff);
Working_view_holder<const_Vector<T, Block2> > w_in(in);
Working_view_holder<Vector<T, Block3> > w_out(out);
Vector<T> kernel = kernel_from_coeff(sym, w_coeff.view);
length_type M = kernel.size(0);
length_type N = in.size(0);
length_type P = out.size(0);
stride_type shift = conv_expected_shift(sup, M);
// expected_P == conv_output_size(sup, M, N, D) == P;
assert(conv_output_size(sup, M, N, D) == P);
Vector<T> sub(M);
// Check result
for (index_type i=0; i<P; ++i)
{
sub = T();
index_type pos = i*D + shift;
if (pos+1 < M)
sub(Domain<1>(0, 1, pos+1)) = w_in.view(Domain<1>(pos, -1, pos+1));
else if (pos >= N)
{
index_type start = pos - N + 1;
sub(Domain<1>(start, 1, M-start)) = w_in.view(Domain<1>(N-1, -1, M-start));
}
else
sub = w_in.view(Domain<1>(pos, -1, M));
w_out.view(i) = dot(kernel, sub);
}
}
void
show(vsip::const_Vector<T, BlockT> vec)
{
using vsip::no_subblock;
using vsip::index_type;
typedef T value_type;
std::cout << "[" << vsip::local_processor() << "] " << "show\n";
if (subblock(vec) != no_subblock)
{
for (index_type li=0; li<vec.local().size(); ++li)
{
index_type gi = global_from_local_index(vec, 0, li);
std::cout << "[" << vsip::local_processor() << "] "
<< li << " gi:" << gi << " = "
<< vec.local().get(li)
<< std::endl;
}
}
else
std::cout << "[" << vsip::local_processor() << "] "
<< "show: no local subblock\n";
}
vsip::Vector<T>
kernel_from_coeff(
vsip::symmetry_type symmetry,
vsip::const_Vector<T, Block> coeff)
{
using vsip::Domain;
using vsip::length_type;
length_type M2 = coeff.size();
length_type M;
if (symmetry == vsip::nonsym)
M = coeff.size();
else if (symmetry == vsip::sym_even_len_odd)
M = 2*coeff.size()-1;
else /* (symmetry == vsip::sym_even_len_even) */
M = 2*coeff.size();
vsip::Vector<T> kernel(M, T());
if (symmetry == vsip::nonsym)
{
kernel = coeff;
}
else if (symmetry == vsip::sym_even_len_odd)
{
kernel(Domain<1>(0, 1, M2)) = coeff;
kernel(Domain<1>(M2, 1, M2-1)) = coeff(Domain<1>(M2-2, -1, M2-1));
}
else /* (symmetry == sym_even_len_even) */
{
kernel(Domain<1>(0, 1, M2)) = coeff;
kernel(Domain<1>(M2, 1, M2)) = coeff(Domain<1>(M2-1, -1, M2));
}
return kernel;
}
// Computes the coefficients for a windowed FIR filter based on
// a given cutoff frequency.
//
// wc The cutoff frequency in radians
//
// w The desired window coefficients (Blackman, Hanning or other)
//
// Defined as
//
// wc
// h(n) = ---- sinc[ wc * (n - M) ] w(n)
// pi
//
// for 0 <= n <= N-1, where N is the length of the input w
//
// Returns a vector of coefficients for a low-pass filter
//
vsip::Vector<vsip::scalar_f>
fir_window(
vsip::scalar_f wc,
vsip::const_Vector<vsip::scalar_f> w)
{
using namespace vsip;
length_type N = w.size();
scalar_f M = (N - 1) / 2.f;
Vector<scalar_f> n = ramp<scalar_f>(0, 1, N);
Vector<scalar_f> wci = wc * (n - M);
Vector<scalar_f> h = wc * sinc(wci) * w / M_PI;
return h;
}
void
corr(
vsip::bias_type bias,
vsip::support_region_type sup,
vsip::const_Vector<T, Block1> ref,
vsip::const_Vector<T, Block2> in,
vsip::Vector<T, Block3> out)
{
using vsip::index_type;
using vsip::length_type;
using vsip::stride_type;
using vsip::Vector;
using vsip::Domain;
using vsip::unbiased;
typedef typename vsip::impl::Scalar_of<T>::type scalar_type;
length_type M = ref.size(0);
length_type N = in.size(0);
length_type P = out.size(0);
length_type expected_P = corr_output_size(sup, M, N);
stride_type shift = expected_shift(sup, M);
assert(expected_P == P);
Vector<T> sub(M);
// compute correlation
for (index_type i=0; i<P; ++i)
{
sub = T();
stride_type pos = static_cast<stride_type>(i) + shift;
scalar_type scale;
if (pos < 0)
{
sub(Domain<1>(-pos, 1, M + pos)) = in(Domain<1>(0, 1, M+pos));
scale = scalar_type(M + pos);
}
else if (pos + M > N)
{
sub(Domain<1>(0, 1, N-pos)) = in(Domain<1>(pos, 1, N-pos));
scale = scalar_type(N - pos);
}
else
{
sub = in(Domain<1>(pos, 1, M));
scale = scalar_type(M);
}
#if VSIP_IMPL_CORR_CORRECT_SAME_SUPPORT_SCALING
#else
if (sup == vsip::support_same)
{
if (i < (M/2)) scale = i + (M+1)/2; // i + ceil(M/2)
else if (i < N - (M/2)) scale = M; // M
else scale = N - 1 + (M+1)/2 - i; // N-1+ceil(M/2)-i
}
#endif
T val = dot(ref, impl_conj(sub));
if (bias == vsip::unbiased)
val /= scale;
out(i) = val;
}
}