602f047fe8
build: replace WIN32 => _WIN32
8 years ago
70ff3804e9
Fix SqueezeNet Halide performance test
8 years ago
a4a47b538c
build: detect Android via '__ANDROID__' macro
...
https://sourceforge.net/p/predef/wiki/OperatingSystems
8 years ago
006966e629
trace: initial support for code trace
8 years ago
e74fed5c84
Catch SkipTestException in performance tests
8 years ago
0e4dde1781
Merge pull request #7872 from alalek:merge-2.4
8 years ago
03bcfe4107
test: fix RNG problem in accuracy tests
9 years ago
71b2409df0
ts: added findDataFile() utility function and SkipTestException
9 years ago
4d84e6efee
fix compiler error when building with gnu-5 toolchain (eg as currently shipped with crystax.org android ndk, and presumably soon with android studio).
...
"ts_perf.cpp ....error: 'errno' was not declared in this scope
int err=errno;"
9 years ago
349d5ba012
--perf_instrument parameter now has int type and 0, 1, 2 modes (1 - simple trees, 2 - expanded trees for functions with same name but different calling address);
...
Maximum depth limit var was added to the instrumentation structure;
Trace names output console output fix: improper tree formatting could happen;
Output in case of error was added;
Custom regions improvements;
Improved timing and weight calculation for parallel regions; New TC (threads counter) value to indicate how many different threads accessed particular node;
parallel_for, warnings fixes and ReturnAddress code from Alexander Alekhin;
9 years ago
10d18a315c
instrumentation: minor refactoring, wrap OpenCL
...
- calculate ticksTotal instead of ticksMean
- local / global width is based on ticksTotal value
- added instrumentation for OpenCL program compilation
- added instrumentation for OpenCL kernel execution
9 years ago
a462d76a95
Implementations instrumentation framework for OpenCV performance system;
9 years ago
12f01b778b
fix perf tests
10 years ago
be507b20cd
fix perf tests
10 years ago
31da833574
ts/ts_perf: fix wrong has() usage
...
`has()` only tests for an argument presence which is always true for
arguments with default values. Use `get<bool>()` to check the value
instead.
10 years ago
6a6d58d389
Adding test support for WINRT
10 years ago
1138a38dbe
Use explicit no-sse flags when SSE is off
...
Also:
- Silence clang warnings about unsupported command line arguments
- Add diagnostic print to calib3d test
- Fixed perf test relative error check
- Fix iOS build problem
10 years ago
8ab6852927
added performance validation check
11 years ago
45958eaabc
Implementation detector and selector for IPP and OpenCL;
...
IPP can be switched on and off on runtime;
Optional implementation collector was added (switched off by default in CMake). Gathers data of implementation used in functions and report this info through performance TS;
TS modifications for implementations control;
11 years ago
05a47d9014
cmd option
11 years ago
a3668f1369
changing default perf strategy from BASIC (old) to SIMPLE (new, faster)
11 years ago
68957b2757
added some new performance tests
11 years ago
32eb38ec98
fixed defects from coverity.com
11 years ago
79eba54a4f
some fixes in perf tests
11 years ago
5abfd40989
added perf test for cv::goodFeaturesToTrack
11 years ago
a7a7b80ff8
added perf test for cv::Moments
11 years ago
d7c22343aa
added perf tests for T-API core functions
12 years ago
7d9150460a
core/ocl: added perf test infra + "Add" perf test
12 years ago
fe01bc21c0
perf: test-based metrics collection strategy
12 years ago
cb4267cb38
ts/perf: int -> enum WarmUpType
12 years ago
f1873bbca1
perf tests: allow to skip performance tests
12 years ago
376cd8f84f
perf: added simple collection strategy
12 years ago
66e4aead97
perf tests: added SANITY_CHECK_NOTHING()
12 years ago
958604a68e
perf test: update error message about relative error
12 years ago
fd88654b45
replaced GPU -> CUDA
12 years ago
cfe4a71dc6
renamed gpu* source to cuda* in core module
12 years ago
e895b7455e
renamed gpu namespace -> cuda
12 years ago
e9a28f66ee
Normalized file endings.
12 years ago
2899d558ac
record cuda gpu name to output report
12 years ago
f87ef77269
removed GPU_/CPU_ prefix from gpu's sanity test
12 years ago
7a104d2793
Added an option to print available implementation variants.
12 years ago
b581f27249
Made perf tests record module name, selected implementation and number of threads.
12 years ago
936236e4b1
Extended the CPU/GPU selection mechanism in performance tests.
...
Now it allows choosing between arbitrary implementation variants.
12 years ago
2dab93c2e8
refactored gpu info structures (TargetArchs and DeviceInfo)
...
now DeviceInfo provides full information about device (from cudaDeviceProp)
12 years ago
4c9773d55f
renamed gpu headers (made it more consistent)
12 years ago
54e7c76d99
fixed compilation errors
12 years ago
4096b54560
use static assertation where suitable
12 years ago
e0ead7b606
fixed assertion failure (vector out of range) for the 'vc10,debug' build ( #2775 )
12 years ago
4f1913ed26
Correct tolerance value for sanity checks with ERROR_RELATIVE
...
Use min/max bounds instead of local value to calculate acceptance threshold.
Threshold based on local values somethimes does not work because cancellation
of big values may produce error bigger than local value.
12 years ago
6385b0f7ed
Fixed a formula to calculate the relative error
12 years ago
Alexander Alekhin
dkurt
Alexander Alekhin
Declan Moran
Pavel Vlasov
Alexander Alekhin
Pavel Rojtberg
Evgeny Agafonchikov
Maksim Shabunin
Ilya Lavrenov
Andrey Pavlenko
Vladislav Vinogradov
Roman Donchenko
marina.kolpakova
Alexey Spizhevoy
Andrey Kamaev
Ivan Korolev