Automatic Differentiation
 
Loading...
Searching...
No Matches
log1m_exp.hpp
Go to the documentation of this file.
1#ifndef STAN_MATH_OPENCL_REV_LOG1M_EXP_HPP
2#define STAN_MATH_OPENCL_REV_LOG1M_EXP_HPP
3#ifdef STAN_OPENCL
4
8
9namespace stan {
10namespace math {
11
18template <typename T,
19 require_all_kernel_expressions_and_none_scalar_t<T>* = nullptr>
21 return make_callback_var(log1m_exp(A.val()),
22 [A](vari_value<matrix_cl<double>>& res) mutable {
23 A.adj() -= elt_divide(res.adj(), expm1(-A.val()));
24 });
25}
26
27} // namespace math
28} // namespace stan
29
30#endif
31#endif
Represents an arithmetic matrix on the OpenCL device.
Definition matrix_cl.hpp:47
fvar< T > log1m_exp(const fvar< T > &x)
Return the natural logarithm of one minus the exponentiation of the specified argument.
Definition log1m_exp.hpp:23
var_value< plain_type_t< T > > make_callback_var(T &&value, F &&functor)
Creates a new var initialized with a callback_vari with a given value and reverse-pass callback funct...
The lgamma implementation in stan-math is based on either the reentrant safe lgamma_r implementation ...