Computes a to the power b for elements of two vectors, where the elements of vector argument a are all non-negative.
Buffer API:
void powr(queue& exec_queue, int64_t n, buffer<T,1>& a, buffer<T,1>& b, buffer<T,1>& y, uint64_t mode = mode::not_defined, error_handler<T> errhandler = {} );
USM API:
event powr(queue& exec_queue, int64_t n, T* a, T* b, T* y, vector_class<event>* depends, uint64_t mode = mode::not_defined, error_handler<T> errhandler = {} );
powr supports the following precisions and devices:
T | Devices supported |
---|---|
float | Host, CPU, and GPU |
double | Host, CPU, and GPU |
The powr(a,b) function raises each element of vector a by the corresponding element of vector b. The elements of a are all nonnegative (ai≥ 0).
Data Type | Threshold Limitations on Input Parameters |
---|---|
single precision | ai < (FLT_MAX)1/bi |
double precision | ai < (DBL_MAX)1/bi |
Special values and VM Error Status treatment for v?Powr function are the same as for pow, unless otherwise indicated in this table:
Argument 1 | Argument 2 | Result | Error Code |
---|---|---|---|
a < 0 | any value b | NAN | status::errdom |
0 < a < ∞ | ±0 | 1 | |
±0 | -∞ < b < 0 | +∞ | |
±0 | -∞ | +∞ | |
±0 | b > 0 | +0 | |
1 | -∞ < b < ∞ | 1 | |
±0 | ±0 | NAN | |
+∞ | ±0 | NAN | |
1 | +∞ | NAN | |
a≥ 0 | NAN | NAN | |
NAN | any value b | NAN | |
0 < a <1 | -∞ | +∞ | |
a > 1 | -∞ | +0 | |
0 ≤a < 1 | +∞ | +0 | |
a > 1 | +∞ | +∞ | |
+∞ | b < +0 | +0 | |
+∞ | b > +0 | +∞ | |
QNAN | QNAN | QNAN | status::errdom |
QNAN | SNAN | QNAN | status::errdom |
SNAN | QNAN | QNAN | status::errdom |
SNAN | SNAN | QNAN | status::errdom |
Buffer API:
The queue where the routine should be executed.
Specifies the number of elements to be calculated.
The buffer a containing 1st input vector of size n.
The buffer b containing 2nd input vector of size n.
Overrides the global VM mode setting for this function call. See set_mode function for possible values and their description. This is an optional parameter. The default value is mode::not_defined.
Sets local error handling mode for this function call. See the create_error_handler function for arguments and their descriptions. This is an optional parameter. The local error handler is disabled by default.
USM API:
The queue where the routine should be executed.
Specifies the number of elements to be calculated.
Pointer a to the 1st input vector of size n.
Pointer b to the 2nd input vector of size n.
Vector of dependent events (to wait for input data to be ready).
Overrides the global VM mode setting for this function call. See the set_mode function for possible values and their description. This is an optional parameter. The default value is mode::not_defined.
Sets local error handling mode for this function call. See the create_error_handler function for arguments and their descriptions. This is an optional parameter. The local error handler is disabled by default.
Buffer API:
The buffer y containing the output vector of size n.
USM API:
Pointer y to the output vector of size n.
Function end event.
An example of how to use powrcan be found in the Intel® oneMKL installation directory, under:
examples/sycl/vml/vpowr.cpp