| COMPMID-2407: Add (logistic and tanh) activation support for QSYMM16 for NEON |
|
More...
|
over 7 years ago
|
| COMPMID-2413: Add QSYMM16 support for PixelWiseMultiplication for CL |
|
More...
|
over 7 years ago
|
| COMPMID-2072: Use a constexpr for num_elems_processed_per_iteration where possible |
|
More...
|
over 7 years ago
|
| COMPMID-2336: Fix argument type for epsilon in CLL2NormalizeLayer. |
|
More...
|
over 7 years ago
|
| COMPMID-2412: Add QSYMM16 support for ElementwiseAddition for CL |
|
More...
|
over 7 years ago
|
| COMPMID-2079: Print positional argument help |
|
More...
|
over 7 years ago
|
| COMPMID-2392: Add documentation for import_memory interface |
|
More...
|
over 7 years ago
|
| COMPMID-2401: Fix CLGemmLowp macro expansion on no-dot platforms |
|
More...
|
over 7 years ago
|
| COMPMID-2092: Refactoring interface for the deconvolution kernels (NEON/CL) |
|
More...
|
over 7 years ago
|
| COMPMID-2245: Extend NEFuseBatchNormalization to support DepthwiseConvolution weights |
|
More...
|
over 7 years ago
|
| COMPMID-2408: Add QSYMM16 support for ElementwiseAddition for NEON |
|
More...
|
over 7 years ago
|
| COMPMID-2053: Fuse bias addition with CLGEMMMatrixMultiplyReshapedKernel |
|
More...
|
over 7 years ago
|
| COMPMID-2343: Add layer normalization support in NELSTMLayer |
|
More...
|
over 7 years ago
|
| COMPMID-2406: Create a new GEMMLowpQuantizeDownInt32ToInt16ScaleKernel for NEON |
|
More...
|
over 7 years ago
|
| COMPMID-2185: Implement direct convolution 9x9 (NHWC) on NEON |
|
More...
|
over 7 years ago
|
| COMPMID-2405: Add lws_hint to CLGEMMLowpMatrixAReductionKernel |
|
More...
|
over 7 years ago
|
| COMPMID-2336: Fix armv8.2-a build on Android |
|
More...
|
over 7 years ago
|
| COMPMID-2387: Add support for NEMeanStdDevNormalizationLayer |
|
More...
|
over 7 years ago
|
| COMPMID-2336: Remove unused variable in PRelu |
|
More...
|
over 7 years ago
|
| COMPMID-2336: Enable implicit fallthrough warnings |
|
More...
|
over 7 years ago
|
| COMPMID-2376: (Nightly) NEON ArgMinMax QASYMM8 |
|
More...
|
over 7 years ago
|
| COMPMID-2380: Create utility functions for is_one and is_zero with float |
|
More...
|
over 7 years ago
|
| COMPMID-2398: Add test for CLFuseBatchNormalizationLayer |
|
More...
|
over 7 years ago
|
| COMPMID-2336: Account for padding in NEIm2ColKernel for NHWC. |
|
More...
|
over 7 years ago
|
| COMPMID-2397: 64bit bare metal build failure |
|
More...
|
over 7 years ago
|
| COMPMID-2393: Fix QASYMM8 activation failures. |
|
More...
|
over 7 years ago
|
| COMPMID-2394: (Nightly) Clang-tidy errors |
|
More...
|
over 7 years ago
|
| COMPMID-2391: Created optical flow example |
|
More...
|
over 7 years ago
|
| COMPMID-2336: Fix warnings present in latest gcc |
|
More...
|
over 7 years ago
|
| COMPMID-2336: Fix armv7a build on linux |
|
More...
|
over 7 years ago
|
