- 4c6bd51 COMPMID-3280: Make all ML primitives for CL use the new interface - Part 1 by Manuel Bottini · 4 years, 3 months ago
- f6ffe99 COMPMID-3349: Fixed performance regression Mali-G71 by Gian Marco Iodice · 4 years, 3 months ago
- 5a4fe19 COMPMID-3112: Reworking heuristic for CLGEMM - part1 by Gian Marco Iodice · 4 years, 4 months ago
- 11d4918 COMPMID-3279: Create CLCompiler interface by Michalis Spyrou · 4 years, 4 months ago
- f64d336 COMPMID-3236: Extend CLGEMMLowpReduction kernels to multiply by a scalar value by Michele Di Giorgio · 4 years, 4 months ago
- fedefc3 COMPMID-2765 Add support for QASYMM8_SIGNED in NEDeconvolutionLayer by Luca Foschiani · 4 years, 5 months ago
- 8b2a7d3 COMPMID-3101 Fuse activation with floating point elementwise operation layers in CL by Giorgio Arena · 4 years, 5 months ago
- cc1f6c9 MLCE-166: Add support for extracting indices in NEPoolingLayer 2x2 NCHW by morgolock · 4 years, 4 months ago
- afc630f COMPMID-3069: Resolve Bfloat16 validation issues. by Georgios Pinitas · 4 years, 4 months ago
- 4b86953 COMPMID-2966 Add support for QASYMM8_SIGNED in NEGEMMLowpQuantizeDownInt32ToUint8ScaleKernel by Luca Foschiani · 4 years, 5 months ago
- 1b14c75 COMPMID-2968: Add support for QASYMM8_SIGNED in CLGEMMLowpQuantizeDownInt32ToUint8ScaleByFloatKernel by Sheri Zhang · 4 years, 4 months ago
- 2e5fd63 COMPMID-3098 Fuse Relu and Bounded Relu with FullyConnected NEON by SiCongLi · 4 years, 5 months ago
- 689c968 COMPMID-2967 Add support for QASYMM8_SIGNED in CLGEMMLowpQuantizeDownInt32ToUint8ScaleKernel by Luca Foschiani · 4 years, 5 months ago
- d9a623a COMPMID-3274: Add support for QASYMM8_SIGNED in CPPBoxWithNonMaximaSuppressionLimit by Sheri Zhang · 4 years, 4 months ago
- 95f8089 COMPMID-3273: Add support for QASYMM8_SIGNED in CPPDetectionPostProcessLayer by Sheri Zhang · 4 years, 4 months ago
- a602f03 COMPMID-3237: Extend GEMMLowpReduction kernels to multiply reductions by a scalar value by Michele Di Giorgio · 4 years, 4 months ago
- c7b183a COMPMID-3160: Add Bfloat16 support in NEGEMMConvolutionLayer by Georgios Pinitas · 4 years, 4 months ago
- 9204646 COMPMID-3069: Remove unnecessary uses of ToolchainSupport.h by Matthew Bentham · 4 years, 4 months ago
- 758b5ba COMPMID-3069: Improve build time by splitting up ToolchainSupport.h by Matthew Bentham · 4 years, 5 months ago
- 0cdbda5 COMPMID-2789: Add support for QASYMM8_SIGNED in CLGEMMDeconvolutionLayer by Sheri Zhang · 4 years, 5 months ago
- b54ba28 COMPMID-2847: Fuse output stage in GEMMLowpMatrixMultiplyReshapedOnlyRHS by Michele Di Giorgio · 4 years, 6 months ago
- b467021 COMPMID-1181: Support for tracing configuration by Anthony Barbier · 6 years ago
- e8291ac COMPMID-3152: Initial Bfloat16 support by Georgios Pinitas · 4 years, 5 months ago
- 773e43c COMPMID-3069: Add NERNN missing assert by Michalis Spyrou · 4 years, 5 months ago
- adb3291 COMPMID-3100 Fuse bias addition with fully connected layer NEON by SiCong Li · 4 years, 5 months ago
- 1856ff7 COMPMID-3097 Fuse activation with fully connected layer CL by Giorgio Arena · 4 years, 5 months ago
- 3c4bf0c COMPMID-3234 CLDirectConvolutionLayer QASYMM8 NHWC mismatches by Giorgio Arena · 4 years, 5 months ago
- 0b192e8 COMPMID-3067: Address gcc9 failures by Georgios Pinitas · 4 years, 5 months ago
- 0ef6032 COMPMID-2788: Add support for QASYMM8_SIGNED in CLDirectDeconvolutionLayer by Sheri Zhang · 4 years, 5 months ago
- 12f2b8c COMPMID-3202: Add support for Valhall architecture in GEMM by Gian Marco Iodice · 4 years, 5 months ago
- ee939fb COMPMID-2774: Add support for QASYMM8_SIGNED in NEReductionOperation, NEReduceMean and NEArgMinMaxLayer by Luca Foschiani · 4 years, 6 months ago
- 695ad69 COMPMID-2964 ~10-15% Regression of DeepSpeech v1 F16 on hikey960 on G71 (CL) by Giorgio Arena · 4 years, 5 months ago
- bbe4458 COMPMID-3084: Fix padding value for Reduce MAX/MIN by Sang-Hoon Park · 4 years, 5 months ago
- 0082c36 Silence warnings related to "control reaches end of non-void function" by Matthew Bentham · 4 years, 6 months ago
- 0b18d97 COMPMID-2762: Add support for QASYMM8_SIGNED in CLReductionOperation and CLReduceMean by Michalis Spyrou · 4 years, 6 months ago
- 55a687d COMPMID-2927: Add support for mixed precision in CLInstanceNormalizationLayer by Georgios Pinitas · 4 years, 6 months ago
- 47fd61f COMPMID-2819: Remove use of std::bind1st as is deprecated since C++11 by Georgios Pinitas · 4 years, 6 months ago
- 11fedda COMPMID-2985 add data_layout to PoolingLayerInfo by Sang-Hoon Park · 4 years, 6 months ago
- b4bb827 COMPMID-2772: Add support for QASYMM8_SIGNED in NEPoolingLayer by Manuel Bottini · 4 years, 7 months ago
- 71ac903 COMPMID-2923 Integrate arm_gemm per channel quantization by Michalis Spyrou · 4 years, 8 months ago
- 19bd412 COMPMID-3022 Fix NEON QASYMM8 Reduction Operation by Michalis Spyrou · 4 years, 6 months ago
- 74ef1db COMPMID-2819: Fix const interface of IMemoryRegion. by Georgios Pinitas · 4 years, 6 months ago
- 0cb3da6 COMPMID-2985 change PoolingLayerInfo to struct by Sang-Hoon Park · 4 years, 6 months ago
- 0a958cb COMPMID-3013: Transpose Convolution 2D VTS/CTS failures on CL by Giuseppe Rossini · 4 years, 6 months ago
- b66aa3b COMPMID-2759 add support for QASYMM8_SIGNED to CLFullyConnectedLayer by Sang-Hoon Park · 4 years, 6 months ago
- cbbed28 COMPMID-2991: Add support for QASYMM8_SIGNED in CL kernels/functions - part 2 by Michele Di Giorgio · 4 years, 7 months ago
- 4715cf9 COMPMID-2760: add support for QASYMM8_SIGNED to CLGEMMConvolutionLayer by Sang-Hoon Park · 4 years, 6 months ago
- 959c26d COMPMID-2790: Add support for QASYMM8_SIGNED in CLGEMMLowpMatrixMultiplyCore by Manuel Bottini · 4 years, 8 months ago
- 9c70037 COMPMID-2769: Add support for QASYMM8_SIGNED in NEFullyConnectedLayer by Michele Di Giorgio · 4 years, 6 months ago
- ab709a0 COMPMID-2727: Add support for split sizes in NESplit by Kurtis Charnock · 4 years, 8 months ago
- ec00da1 COMPMID-2728: Add support for split sizes in CLSplit by Kurtis Charnock · 4 years, 8 months ago
- 13ec5f0 COMPMID-2800: Add support for QASYMM8_SIGNED in NEDepthwiseConvolutionLayer3x3Kernel by Michele Di Giorgio · 4 years, 7 months ago
- 781d727 COMPMID-2994: Add support QASYMM8_SIGNED in NEElementwiseMin by morgolock · 4 years, 6 months ago
- 5b7fd87 MLCE-139 add align_corners parameter handling to CLScale by Sang-Hoon Park · 4 years, 6 months ago
- 8481d83 COMPMID-2753: Add support for QASYMM8_SIGNED in CL kernels/functions by Manuel Bottini · 4 years, 7 months ago
- a359805 COMPMID-2994: Add support QASYMM8_SIGNED in NEElementwiseMax by morgolock · 4 years, 7 months ago
- bb123bd MLCE-139 add align_corners parameter to NEScale by Sang-Hoon Park · 4 years, 7 months ago
- 4536193 COMPMID-2801: Add support for QASYMM8_SIGNED in NEDirectConvolutionLayerOutputStageKernel by Michele Di Giorgio · 4 years, 7 months ago
- 807ce59 COMPMID-2999: Nighliy Build Failure [655] - Bare metal build for armv7 by Michele Di Giorgio · 4 years, 7 months ago
- 5255672 COMPMID-2998: NEON/GEMMConvolutionLayer/Quantized/QASYMM8_SIGNED failing on 32-bit by Michele Di Giorgio · 4 years, 7 months ago
- f29d1b7 COMPMID-2608: Enable quantization with multiplier greater than 1 on NEON by Michele Di Giorgio · 4 years, 9 months ago
- 0779fec COMPMID-2763 [CL] add support for QASYMM8_SIGNED to SoftmaxLayer by Sang-Hoon Park · 4 years, 8 months ago
- e7be8a0 COMPMID-2980 (Nightly) armv7a build failures by Michalis Spyrou · 4 years, 7 months ago
- a0a4ba1 COMPMID-2855: CLReduceMean throws error for invalid configs by Pablo Tello · 4 years, 7 months ago
- 3384356 COMPMID-2754: Add support for QASYMM8_SIGNED in NE kernels/functions. by Georgios Pinitas · 4 years, 7 months ago
- f01201a COMPMID-2305: NEDepthwiseConvolution 3x3: support for QUANT8_PER_CHANNEL_SYMM by Giuseppe Rossini · 4 years, 9 months ago
- 6e1791b COMPMID-2764: Add support for QASYMM8_SIGNED in NEConvolutionLayer. by Georgios Pinitas · 4 years, 8 months ago
- 5cb49dc COMPMID-2962 Fix compiler warnings for g++8 by Michalis Spyrou · 4 years, 8 months ago
- 1f332d4 COMPMID-2794: Add support for QASYMM8_SIGNED in CLGEMMLowpOutputStage by Manuel Bottini · 4 years, 8 months ago
- 7b9998d COMPMID-1816: Use parallel reduction on 0 axis in CL ARG_MIN/ARG_MAX by Manuel Bottini · 4 years, 9 months ago
- f9179d3 COMPMID-2793: Add support for QASYMM8_SIGNED in CLGEMMLowpMatrixMultiplyReshapedOnlyRHSKernel by Michele Di Giorgio · 4 years, 8 months ago
- 14cbfb2 COMPMID-2609: Enable quantization with multiplier greater than 1 on OpenCL by Michele Di Giorgio · 4 years, 9 months ago
- ae6ef7c COMPMID-2763 convert functions and helpers for QASYMM8_SIGNED by Sang-Hoon Park · 4 years, 8 months ago
- 448a81f COMPMID-2805: Add QASYMM8_SIGNED support in NEGEMMLowpOutputStage by Georgios Pinitas · 4 years, 8 months ago
- 646d4e4 COMPMID-2952: NEON failing with 'bus_error' on 32-bit by Georgios Pinitas · 4 years, 8 months ago
- 9397515 COMPMID-2855: NEReduceMean throws error for invalid configs by Pablo Tello · 4 years, 8 months ago
- 581f178 COMPMID-2920: NEInstanceNormalization fails on NHWC validations by Manuel Bottini · 4 years, 8 months ago
- 2963a6f COMPMID-2152: Implement Fill for NEON by Kurtis Charnock · 4 years, 8 months ago
- 9afc41a COMPMID-2151: Implement Fill for CL by Kurtis Charnock · 4 years, 8 months ago
- 6011f24 COMPMID-2853: Fix VGG16 for fp16. by Georgios Pinitas · 4 years, 8 months ago
- 951b8a4 COMPMID-2309 : CLConvolutionLayer: support QUANT8_SYMM_PER_CHANNEL filters by Vidhya Sudhan Loganathan · 4 years, 9 months ago
- 7a669a8 COMPMID-2921 Performance regression on NEON Int8 by Giorgio Arena · 4 years, 8 months ago
- eaa01ab COMPMID-2671 use Signed32 for default output data type of ArgMinMax by Sang-Hoon Park · 4 years, 8 months ago
- 77d4252 COMPMID-2853: VGG16 regression for fp32 by Georgios Pinitas · 4 years, 9 months ago
- 63d4dbd COMPMID-2452: Fix 32-bit per-channel convolution for NEON. by Georgios Pinitas · 4 years, 8 months ago
- 5264b7d COMPMID-2576: Fuse activation in Winograd output transform. by Pablo Tello · 4 years, 9 months ago
- dbdea0d COMPMID-2308: NEConvolutionLayer: support QUANT8_SYMM_PER_CHANNEL filters by Georgios Pinitas · 4 years, 9 months ago
- 0fc2d7a COMPMID-2862: remove forward declaration of ReductionOperation by Sang-Hoon Park · 4 years, 9 months ago
- 7ae80a9 COMPMID-2549: Add GLES Runtime Context interfaces. by Georgios Pinitas · 4 years, 9 months ago
- 62eeb53 COMPMID-2266: [CL] add support for Log Softmax by Sang-Hoon Park · 4 years, 9 months ago
- 44bfc3f COMPMID-1671: Allow fp mixed precision in CLFCLayer. by Georgios Pinitas · 4 years, 9 months ago
- 2697fd8 COMPMID-2707: add keep_dims parameter to Reduction Operation by Sang-Hoon Park · 4 years, 9 months ago
- df4cf57 COMPMID-2306: CLDepthwiseConvolution: support for QUANT8_PER_CHANNEL_SYMM by Michele Di Giorgio · 4 years, 9 months ago
- 8b72199 COMPMID-1889: Fuse bias addition and output stage in CLFCLayer. by Georgios Pinitas · 4 years, 9 months ago
- d985378 COMPMID-2588: Optimize the output detection kernel required by MobileNet-SSD (~27% improvement) by Giuseppe Rossini · 4 years, 9 months ago
- 279814b COMPMID-2481: Add InputQuantizationInfo!=OutputQuantization Info tests for DeconvolutionLayer by Manuel Bottini · 4 years, 9 months ago
- 0726398 COMPMID-2501: Support multiplier > 1 during QASYMM8 requantization for Quantized LSTM by Manuel Bottini · 4 years, 9 months ago
- 05069f0 COMPMID-2515: Merge optimized depthwise convolution to the generic depthwise convolution function by Manuel Bottini · 4 years, 10 months ago
- 48b3ef8 COMPMID-2577: Fuse bias addition and activation in gemm assembly kernels by Georgios Pinitas · 4 years, 9 months ago
- d93e263 COMPMID-2708 NEDepthwiseConvolution Generic: support for QUANT8_PER_CHANNEL_SYMM by Giorgio Arena · 4 years, 9 months ago