| COMPMID-3335: Remove x/y-axis padding from CLGEMMReshapeLHSMatrixKernel |
|
More...
|
about 6 years ago
|
| COMPMID-3339: Patch1: Fix incorrect select casting in im2col nhwc kernels |
|
More...
|
about 6 years ago
|
| MLCE-229: Support for negative shifts in asm kernels |
|
More...
|
about 6 years ago
|
| COMPMID-3700: ArmNN OOB test failure |
|
More...
|
about 6 years ago
|
| COMPMID-3454 Patch1: Relocate data_type_from_name to core/Utils |
|
More...
|
about 6 years ago
|
| COMPMID-3502: Add support of different quantization input/output for ReduceMean |
|
More...
|
over 6 years ago
|
| COMPMID-3683: Fix performance regression on Mali-G76 (Fully connected) COMPMID-3682: Fix performance regression on Mali-G76 (Convolution) |
|
More...
|
over 6 years ago
|
| Added missing parameter num_groups to the validate call of NEConvolutionLayer. |
|
More...
|
over 6 years ago
|
| COMPMID-3339: Patch2: Remove paddings from im2col*_nhwc cl kernel |
|
More...
|
over 6 years ago
|
| COMPMID-3656: Disabled reduce_axis in LOG_SOFTMAX and SOFTMAX |
|
More...
|
over 6 years ago
|
| COMPMID-2479: Extend CLPoolingLayer max pooling to extract indices |
|
More...
|
over 6 years ago
|
| COMPMID-3652 Fix CLFullyConnectedLayer failure on S10 |
|
More...
|
over 6 years ago
|
| COMPMID-3392: Collapse TensorMaps into a single TensorPack |
|
More...
|
over 6 years ago
|
| COMPMID-3526: LOGISTIC support for values bigger than [-40.f,40.f] |
|
More...
|
over 6 years ago
|
| COMPMID-3324: Fix oclgrind warnings |
|
More...
|
over 6 years ago
|
| COMPMID-3324: Remove configuration tests |
|
More...
|
over 6 years ago
|
| COMPMID-3624: CTS failure on Resize quantized in Neon and CL |
|
More...
|
over 6 years ago
|
| COMPMID-2078: Remove legacy TODOs |
|
More...
|
over 6 years ago
|
| Fix EltwiseLayerNode and QuantizationLayerNode |
|
More...
|
over 6 years ago
|
| Add support ElementwiseMax operator in graph API |
|
More...
|
over 6 years ago
|
| COMPMID-3585: Android R while_fib_n_5_quant8 failure on CpuAcc |
|
More...
|
over 6 years ago
|
| COMPMID-3618 Add support for export_to_cl_image_rhs in GEMMTuner.py |
|
More...
|
over 6 years ago
|
| COMPMID-3653 CL GEMM kernel creation error on certain combinations of N and N0 |
|
More...
|
over 6 years ago
|
| COMPMID-3385: Async support to CLArithmetic* kernels/functions Pt.2 |
|
More...
|
over 6 years ago
|
| COMPMID-3385: Async support to CLArithmetic* kernels/functions Pt.1 |
|
More...
|
over 6 years ago
|
| COMPMID-3600: Fix requantization in NEPixelWiseMultiplicationKernel |
|
More...
|
over 6 years ago
|
| COMPMID-3393: Minor tweaks on memory injection interface |
|
More...
|
over 6 years ago
|
| COMPMID-3535: 9x9 Direct convolution support for CL and NHWC |
|
More...
|
over 6 years ago
|
| COMPMID-3326; Update heuristic for GEMMReshaped and GEMMReshapedOnlyRHS |
|
More...
|
over 6 years ago
|
| COMPMID-3532: Nighly tolerance failure on CLDirectConvolutionLayer 9x9 |
|
More...
|
over 6 years ago
|
