|
|
Prev | Next | base_identical |
AD<Base>
,
the type
Base
must support the syntax
b = CppAD::EqualOpSeq(u, v)
u
is identically equal to
v
and it makes no different which one is used.
The arguments
u
and
v
have prototype
const Base& u
const Base& v
b
has prototype
bool b
Base
is a relatively simple type,
the EqualOpSeq function can be defined by
namespace CppAD {
inline Base EqualOpSeq(const Base& u, const Base& v)
{ return u == v; }
}
u
and
v
are not identically equal in the following case
(which CppAD automatically defines EqualOpSeq for):
The type
Base
is
AD<double>
,
x[0] = x[1] = 1.
,
then independent
is used to make
x
the independent
variable vector,
and then
u = x[0]
,
v = x[1]
,
Note that during a future Forward
calculation,
u
and
v
could correspond to different values.
For example, see
adolc EqualOpSeq
.
Base
is a parameter
when used in an
AD<Base>
operation sequence.
It is however still possible for a parameter to change its value.
For example,
the
Base
value
u
is not identically a parameter
equal in the following case
(which CppAD automatically defines IdenticalPar for):
The type
Base
is
AD<double>
,
x[0] = 1.
,
then independent
is used to make
x
the independent
variable vector,
and then
u = x[0]
,
Note that during a future Forward
calculation,
u
could correspond to different values.
u
has prototype
const Base u
v
has prototype
const Base v
b
has prototype
bool b
Base
must support the following functions
(in the CppAD namespace):
| Syntax | Result |
b = IdenticalPar(u)
|
the
Base
value will always be the same
|
b = IdenticalZero(u)
|
u
equals zero and
IdenticalPar(u)
|
b = IdenticalOne(u)
|
u
equals one and
IdenticalPar(u)
|
b = IdenticalEqualPar(u, v)
|
u
equals
v
,
IdenticalPar(u)
and
IdenticalPar(v)
|