| COMPMID-2336: Fix build for v8.2-a architectures |
|
More...
|
over 7 years ago
|
| COMPMID-2457: Investigate DirectConvolution failures |
|
More...
|
over 7 years ago
|
| COMPMID-2432: Fix defects reported by Coverity |
|
More...
|
over 7 years ago
|
| COMPMID-2425: Fix memory leak reported by coverity |
|
More...
|
over 7 years ago
|
| COMPMID-2336: Set the correct tolerance in PixeWiseMultiplication test |
|
More...
|
over 7 years ago
|
| COMPMID-2443: OpenCL kernels cache. |
|
More...
|
over 7 years ago
|
| COMPMID-2193: Improve permute documentation. |
|
More...
|
over 7 years ago
|
| -ftree-vectorize is redundant with -O3 |
|
More...
|
over 7 years ago
|
| COMPMID-2435: Update GPU target definitions |
|
More...
|
over 7 years ago
|
| COMPMID-2433: (Nightly) Bug in CLDequantizationLayer |
|
More...
|
over 7 years ago
|
| COMPMID-2410: Create a new GEMMLowpQuantizeDownInt32ToInt16ScaleKernel for CL |
|
More...
|
over 7 years ago
|
| COMPMID-2336: Fix enable tuning of kernels targeted in COMPUTE-10611 |
|
More...
|
over 7 years ago
|
| COMPMID-2432 Fix defects reported by Coverity |
|
More...
|
over 7 years ago
|
| COMPMID-2234 : Add support for axis 3 in NE/CLConcatenateLayer |
|
More...
|
over 7 years ago
|
| COMPMID-2429: Unexpected exception in destructor |
|
More...
|
over 7 years ago
|
| COMPMID-2427: Remove structurally dead code |
|
More...
|
over 7 years ago
|
| COMPMID-2426: Logical dead code fix |
|
More...
|
over 7 years ago
|
| COMPMID-2428: Missing break statement. |
|
More...
|
over 7 years ago
|
| COMPMID-2178: Update GEMM assembly code. |
|
More...
|
over 7 years ago
|
| COMPMID-2431: Add a comment to max_offset in CLDepthwiseConvolutionLayer3x3NHWCKernel |
|
More...
|
over 7 years ago
|
| COMPMID-2336 Fix FuseBatchNormalizationFixture |
|
More...
|
over 7 years ago
|
| COMPMID-1979: Fuse Activation Function in CLGEMM - part 1 |
|
More...
|
over 7 years ago
|
| COMPMID-2411: Add (logistic and tanh) activation support for QSYMM16 for CL |
|
More...
|
over 7 years ago
|
| COMPMID-2409: Add QSYMM16 support for PixelWiseMultiplication for NEON |
|
More...
|
over 7 years ago
|
| COMPMID-2096: Refactor the CLGEMMLowp function selection (heuristic) |
|
More...
|
over 7 years ago
|
| COMPMID-2156: Optimized dilated convolution for NEON. |
|
More...
|
over 7 years ago
|
| COMPMID-2172: Fuse bias addition with CLGEMMMatrixMultiplyNativeKernel |
|
More...
|
over 7 years ago
|
| COMPMID-2336: Fix Nightly bug NEArithmeticAddition |
|
More...
|
over 7 years ago
|
| COMPMID-2378: Sanitize GEMM configuration for NEON |
|
More...
|
over 7 years ago
|
| COMPMID-2273: Fuse Batch Normalization with Depthwise Convolution layer at graph level (only for CL) |
|
More...
|
over 7 years ago
|
