Age | Commit message (Expand) | Author |
2019-02-07 | COMPMID-1923: Input is ignored if peephole optimization is enabled in | Georgios Pinitas |
2019-02-06 | COMPMID-1915: Deconvolution doesn't work when inner_dimension_top != 0 or inn... | giuros01 |
2019-02-06 | COMPMID-1838: Add 4D softmax support for NEON and achieve parity with CL | Manuel Bottini |
2019-02-04 | COMPMID-1710: Introduce GEMM strategy name in GEMMAssemblyWrapper. | Georgios Pinitas |
2019-02-04 | COMPMID-1823: Increase scheduling granularity in NEGemmInterleaved. | Georgios Pinitas |
2019-02-01 | COMPMID-1710: Re-enable winograd on A53. | Georgios Pinitas |
2019-01-25 | COMPMID-1710: Define missed specializations for comparisons | Georgios Pinitas |
2019-01-23 | COMPMID-1760: NEON: Implement Prod | Manuel Bottini |
2019-01-21 | COMPMID-1768: NEON: Implement RealDiv | George Wort |
2019-01-21 | COMPMID-1753: NEON: Implement Less, Greater, GreaterEqual, Equal, Not Equal | George Wort |
2019-01-21 | COMPMID-1763 : NEON: Implement Gather | John Kesapides |
2019-01-18 | COMPMID-1867: Add NEON/SVE GEMM Hybrid kernels. | Georgios Pinitas |
2019-01-18 | COMPMID-1814 : NEScale add support for TOP_LEFT and QASYMM8 | Vidhya Sudhan Loganathan |
2019-01-17 | COMPMID-1891: Fixes NELSTM padding issues. | Georgios Pinitas |
2019-01-15 | COMPMID-1724: CL Implement Prod fix | Manuel Bottini |
2019-01-14 | COMPMID-1809: Remove padding in NEGEMMConvolutionLayer 64-bit path. | Georgios Pinitas |
2019-01-14 | COMPMID-1772: Implement PadV2 for NEON | Georgios Pinitas |
2019-01-14 | COMPMID-1758: NEON: Implement Range | Manuel Bottini |
2019-01-14 | COMPMID-1724: CL Implement Prod | Manuel Bottini |
2019-01-11 | COMPMID-1677: Change ROIPooling layer interface to accept ROIs as tensors | Manuel Bottini |
2019-01-11 | COMPMID-1761: NEON: Implement Pack | Isabella Gottardi |
2019-01-10 | COMPMID-1764 NEON: Implement ArgMax/ArgMin | Michalis Spyrou |
2019-01-08 | COMPMID-1759 NEON: Implement Reverse | Michalis Spyrou |
2019-01-08 | COMPMID-1865 NEReduceMean fails on shape validation | Michalis Spyrou |
2019-01-08 | COMPMID-1756: NEON: Implement RSqrt, Exp | George Wort |
2019-01-02 | COMPMID-1769: Add support for NEON StridedSlice,Split,Slice,Unstack | Georgios Pinitas |
2019-01-02 | COMPMID-1767: NEON: Implement Where/Select | George Wort |
2018-12-17 | COMPMID-1754: NEON: Implement Maximum, Minumum, SquaredDifference | giuros01 |
2018-12-13 | COMPMID-1741: Implement NEFuseBatchNormalizationKernel | giuros01 |
2018-12-11 | COMPMID-1497: Add support for interleaved B reshaping in gemm_interleaved | Anthony Barbier |
2018-12-05 | COMPMID-1757: NEON: Implement Tile | giuros01 |
2018-11-30 | COMPMID-1815 Remove templates from NEFillborderKernel and create INESimpleFun... | Michalis Spyrou |
2018-11-23 | COMPMID-1451: Fix MemoryManager usage. | Georgios Pinitas |
2018-11-22 | COMPMID-1645 NEL2Normalization for FP32/FP16 & NHWC | Michalis Spyrou |
2018-11-19 | COMPMID-1644: NEDepthwiseConvolution for FP16 NHWC | Georgios Pinitas |
2018-11-16 | COMPMID-1461 SSD support: Create NEON PriorBox | Michalis Spyrou |
2018-11-15 | COMPMID-1783: NEGEMMConvolutionLayer fails for QASYMM8 on 32bit | Georgios Pinitas |
2018-11-15 | COMPMID-1451: Fix the shape of scratch_buffer in case of CIFG | Georgios Pinitas |
2018-11-13 | COMPMID-1751: Remove output_3d_depth from NEGEMMLowpQuantizeDownInt32ToUint8S... | Georgios Pinitas |
2018-11-09 | COMPMID-1626: Fixed VGG 16/19 bad_alloc failure. | Pablo Tello |
2018-11-09 | COMPMID-1451: Dispatch fp16 gemms through the function interface. | Georgios Pinitas |
2018-11-08 | COMPMID-1451: Fix fused activation in GEMMConvolutionLayer | Georgios Pinitas |
2018-11-08 | COMPMID-1736: Fixed out-of-bound write in CLIm2Col | Gian Marco Iodice |
2018-11-08 | COMPMID-1675: Add SVE support | Georgios Pinitas |
2018-11-08 | COMPMID-1579: Add support for ChannelShuffle operator in NEON | Georgios Pinitas |
2018-11-08 | COMPMID-1773: Replaced Gemm wrapper for NEGEMM in NEWinograd | Pablo Tello |
2018-11-02 | COMPMID-1701: small cleanup NEWinogradConvolutionLayer. | Pablo Tello |
2018-11-02 | COMPMID-1413 - Improve the performance of GEMMLowp with 8 bit dot product on ... | Gian Marco Iodice |
2018-11-02 | COMPMID-1451: Remove invalid bias check in NEDeconvolutionLayer | Georgios Pinitas |
2018-11-02 | COMPMID-1451: Fuse activation in DepthwiseConvolution. | Georgios Pinitas |