- 30 7月, 2018 4 次提交
-
-
由 Alexander Alekhin 提交于
-
由 miaow1988 提交于
Changed the type of variable *r* from int to size_t. This change makes sure that a valid result of std::max(r + delta, (r*3+1)/2) can be passed into the reserve function.
-
由 Alexander Alekhin 提交于
-
由 Sayed Adel 提交于
-
- 29 7月, 2018 1 次提交
-
-
由 Alexander Alekhin 提交于
-
- 28 7月, 2018 2 次提交
-
-
由 Alexander Alekhin 提交于
-
由 Maksim Shabunin 提交于
-
- 27 7月, 2018 2 次提交
-
-
由 Alexander Alekhin 提交于
-
由 Maksim Shabunin 提交于
-
- 26 7月, 2018 8 次提交
-
-
由 Alexander Alekhin 提交于
-
由 Alexander Alekhin 提交于
-
由 Alexander Alekhin 提交于
-
由 Maksim Shabunin 提交于
-
由 Triplesalt 提交于
Squashed : ReLU operand order tests.
-
由 Vadim Pisarevsky 提交于
-
由 Vadim Pisarevsky 提交于
* 1. changed static const __m128/256 to const __m128/256 to avoid wierd instructions and calls inserted by compiler. 2. added universal intrinsics that wrap MOVNTPS and other such (non-temporary or "no cache" store) instructions. v_store_interleave() and v_store() got respective flags/overloaded variants 3. rewrote split & merge to use the "no cache" store instructions. It resulted in dramatic performance improvement when processing big arrays * hopefully, fixed some test failures where 4-channel v_store_interleave() is used * added missing implementation of the new universal intrinsics (v_store_aligned_nocache() etc.) * fixed silly typo in the new intrinsics in intrin_vsx.hpp * still trying to fix VSX compiler errors * still trying to fix VSX compiler errors * still trying to fix VSX compiler errors * still trying to fix VSX compiler errors
-
由 Dmitry Kurtaev 提交于
Enable Torch layers tests with Intel's Inference Engine backend.
-
- 25 7月, 2018 6 次提交
-
-
由 Alexander Alekhin 提交于
but execute tested functions in Layer_Test_Halide/Convolution.Accuracy
-
由 Alexander Alekhin 提交于
instead of 'int'
-
由 Alexander Alekhin 提交于
-
由 Alexander Alekhin 提交于
-
由 Alexander Alekhin 提交于
-
由 Alexander Alekhin 提交于
-
- 24 7月, 2018 11 次提交
-
-
由 Vadim Pisarevsky 提交于
* fixed/updated v_load_deinterleave and v_store_interleave intrinsics; modified split() and merge() functions to use those intrinsics * fixed a few compile errors and bug in v_load_deinterleave(ptr, v_uint32x4& a, v_uint32x4& b) * fixed few more compile errors
-
由 Maksim Shabunin 提交于
-
由 Alexander Alekhin 提交于
-
由 Alexander Alekhin 提交于
-
由 Alexander Alekhin 提交于
-
由 Alexander Alekhin 提交于
improve error checks
-
由 Alexander Alekhin 提交于
-
由 Alexander Alekhin 提交于
-
由 Alexander Alekhin 提交于
-
由 Dmitry Kurtaev 提交于
-
由 Dmitry Kurtaev 提交于
-
- 23 7月, 2018 2 次提交
-
-
由 Maksim Shabunin 提交于
-
由 Alexander Alekhin 提交于
-
- 22 7月, 2018 1 次提交
-
-
由 Paul92 提交于
Mser sample improvments (#12032) * Fixed bug in detect_mser sample Wrong number of colors used to generate the synthetic images * Formatting improvements * Using safer casts * Improved readability of legend generation * Various readability fixes in detect_mser sample
-
- 21 7月, 2018 3 次提交
-
-
由 Alexander Alekhin 提交于
-
由 Alexander Alekhin 提交于
-
由 Teng Yiliang 提交于
the static variables will cause race-condition when operating in multithread scenarios. Signed-off-by: NTeng Yiliang <ylteng@outlook.com> Signed-off-by: NTeng Yiliang <yiliang.teng@weimob.com>
-