%PDF- %PDF-
| Direktori : /proc/thread-self/root/lib/python3/dist-packages/pythran/pythonic/numpy/random/ |
| Current File : //proc/thread-self/root/lib/python3/dist-packages/pythran/pythonic/numpy/random/laplace.hpp |
#ifndef PYTHONIC_NUMPY_RANDOM_LAPLACE_HPP
#define PYTHONIC_NUMPY_RANDOM_LAPLACE_HPP
#include "pythonic/include/numpy/random/laplace.hpp"
#include "pythonic/include/numpy/random/generator.hpp"
#include "pythonic/types/ndarray.hpp"
#include "pythonic/types/NoneType.hpp"
#include "pythonic/types/tuple.hpp"
#include "pythonic/utils/functor.hpp"
#include <random>
#include <algorithm>
PYTHONIC_NS_BEGIN
namespace numpy
{
namespace random
{
template <class pS>
types::ndarray<double, pS> laplace(double loc, double scale,
pS const &shape)
{
types::ndarray<double, pS> result{shape, types::none_type()};
std::generate(result.fbegin(), result.fend(),
[&]() { return laplace(loc, scale); });
return result;
}
auto laplace(double loc, double scale, long size)
-> decltype(laplace(loc, scale, types::array<long, 1>{{size}}))
{
return laplace(loc, scale, types::array<long, 1>{{size}});
}
double laplace(double loc, double scale, types::none_type d)
{
double U =
std::uniform_real_distribution<double>{0., 1.}(details::generator);
if (U >= 0.5) {
U = loc - scale * xsimd::log(2.0 - U - U);
} else if (U > 0.0) {
U = loc + scale * xsimd::log(U + U);
} else {
U = laplace(loc, scale);
}
return U;
}
}
}
PYTHONIC_NS_END
#endif