1. 2ffc5e6 AVX512 versions of QS8 DWCONV microkernels by Marat Dukhan · 4 years, 1 month ago
  2. 70a9618 End-to-end QS8 MobileNet v2 benchmark by Marat Dukhan · 4 years, 1 month ago
  3. ff20948 QS8 version of ND ADD operator by Marat Dukhan · 4 years, 1 month ago
  4. 5df27f8 WAsm SIMD versions of QS8 VADD[C] microkernels by Marat Dukhan · 4 years, 1 month ago
  5. ba7b279 NEON variants of QS8 VADD[C] microkernels by Marat Dukhan · 4 years, 1 month ago
  6. 95caee5 Test zero_point and scale parameters in QS8 VADD[C] microkernels by Marat Dukhan · 4 years, 1 month ago
  7. 9c7308f vbinary microkernels unrolled to x8 for scalar and web assembly and x16 web assembly simd by Frank Barchard · 4 years, 1 month ago
  8. a635a29 WASM and WASMSIMD test ifdef improved. by Frank Barchard · 4 years, 1 month ago
  9. 37297a6 F32-RELU unrolled more for improved performance on Web Assembly by Frank Barchard · 4 years, 1 month ago
  10. 4422f10 FP16 hswish fix cast fix for improved GCC compatability. by Frank Barchard · 4 years, 1 month ago
  11. 12c5777 Optimization: 2x partial unroll to load 8 contiguous bytes. by Benoit Jacob · 4 years, 1 month ago
  12. 919328c Improve compatibility with -fno-lax-vector-conversions by Marat Dukhan · 4 years, 1 month ago
  13. 7359463 Fix incompatibilities with ARM GCC by Marat Dukhan · 4 years, 1 month ago
  14. c2146cc Added support for specialized batch_size == 1 convolutions. by Fabio Riccardi · 4 years, 1 month ago
  15. 62d752e Function macro for WASM assembly by Frank Barchard · 4 years, 1 month ago
  16. 9bd894b Relu wasm assembly use if statements instead of block for readability. by Frank Barchard · 4 years, 1 month ago
  17. bcedc08 Move WAsm assembly source into its own list of sources by Frank Barchard · 4 years, 1 month ago
  18. 9f5ccb3 f32 relu microkernel WASM assembly x2 and x4 unrolled loops by Frank Barchard · 4 years, 1 month ago
  19. 6fe0b53 add wasm32 to common build scripts to define macros for unittests and benchmarks by Frank Barchard · 4 years, 1 month ago
  20. a05487f Add xnn_qs8_igemm_minmax_ukernel_${MR}x${NR}c4__neondot (ARMv8.2+dotprod). by Benoit Jacob · 4 years, 1 month ago
  21. 799ac75 Use more verbose compiler flags for NEON DOT ISA on AArch32 by Marat Dukhan · 4 years, 2 months ago
  22. a964473 Add xnn_qs8_gemm_minmax_ukernel_${MR}x${NR}c4__neondot (ARMv8.2+dotprod). by Benoit Jacob · 4 years, 2 months ago
  23. 0af63ab Include polyfills for intrinsics in QS8 AVX512 GEMM/IGEMM microkernels by Marat Dukhan · 4 years, 2 months ago
  24. d91624e f32 relu microkernel WASM assembly optimized implementation by Frank Barchard · 4 years, 2 months ago
  25. 2a18f7e Polyfill _mm512_set_epi8 for older gcc by Marat Dukhan · 4 years, 2 months ago
  26. 0270d9f QS8 VADDC microkernels in SSE2 and SSE4.1 implementations by Marat Dukhan · 4 years, 2 months ago
  27. 8432486 Fix CMake build by Marat Dukhan · 4 years, 2 months ago
  28. f28cddf Initialize QS8 microkernels in ARM/ARM64 builds by Marat Dukhan · 4 years, 2 months ago
  29. 281262d NEON variant of QS8 GAVGPOOL microkernel by Marat Dukhan · 4 years, 2 months ago
  30. 023bcf9 NEON variant of QS8 DWCONV microkernels by Marat Dukhan · 4 years, 2 months ago
  31. d9f3ad4 QS8 ADD microkernels in SSE2 and SSE4.1 implementations by Marat Dukhan · 4 years, 2 months ago
  32. bb00b1d AVX512 variants of QS8 GEMM and IGEMM microkernels by Marat Dukhan · 4 years, 2 months ago
  33. facecc5 Fix fusion of Constant Pad with asymmetric padding by Marat Dukhan · 4 years, 2 months ago
  34. f124e88 Polyfill _mm_loadu_si32 and _mm_storeu_si32 intrinsics by Marat Dukhan · 4 years, 2 months ago
  35. 4de076d Fix xnn_update_qs8_avgpool_params on WAsm SIMD by Marat Dukhan · 4 years, 2 months ago
  36. 674778d Add binary op microkernels with RELU activation by Frank Barchard · 4 years, 2 months ago
  37. 75215d8 Enable XOP versions of GEMM/IGEMM microkernels by Marat Dukhan · 4 years, 2 months ago
  38. c15aa4e Remove XOP variants of QS8 DWCONV by Marat Dukhan · 4 years, 2 months ago
  39. 5b3af47 Re-generate QS8 and QU8 microkernels from templates by Marat Dukhan · 4 years, 2 months ago
  40. 453f4b8 Fix CMake build by Marat Dukhan · 4 years, 2 months ago
  41. c9c320e CMake build fix by Frank Barchard · 4 years, 2 months ago
  42. b33fc0e Add xnn_q{u,s}8_gemm_minmax_ukernel_MRxNRc4__scalar by Benoit Jacob · 4 years, 2 months ago
  43. 9e0b539 QS8 variant of NWC Global Average Pooling operator by Marat Dukhan · 4 years, 2 months ago
  44. 4013552 AVX2 versions of QS8 DWCONV microkernels using 16-bit multiplication by Marat Dukhan · 4 years, 2 months ago
  45. 54e95a0 Unify implementation of NWC Global Average Pooling across data types by Marat Dukhan · 4 years, 2 months ago
  46. b5e3d17 Multipass QS8 GAVGPOOL microkernel in WAsm SIMD implementation by Marat Dukhan · 4 years, 2 months ago
  47. ef45180 Unipass QS8 GAVGPOOL microkernel in WAsm SIMD implementation by Marat Dukhan · 4 years, 2 months ago
  48. 159688f Multipass QS8 GAVGPOOL microkernels in SSE2/SSSE3/SSE4.1 implementations by Marat Dukhan · 4 years, 2 months ago
  49. 4ed53f4 Unipass QS8 GAVGPOOL microkernels in SSE2/SSSE3/SSE4.1 implementations by Marat Dukhan · 4 years, 2 months ago
  50. 07e5040 Initialize QS8 microkernels for WAsm SIMD by Marat Dukhan · 4 years, 2 months ago
  51. ab67142 Benchmark ARM NEON versions of QS8 GEMM microkernels by Marat Dukhan · 4 years, 2 months ago
  52. cc8f34c WAsm SIMD variants of QS8 DWCONV microkernels by Marat Dukhan · 4 years, 2 months ago
  53. 27203da WAsm SIMD versions of QS8 GEMM and IGEMM microkernels by Marat Dukhan · 4 years, 2 months ago
  54. b1cbbba Add ADE20k parameters to Softmax benchmark by Marat Dukhan · 4 years, 2 months ago
  55. 23848db Reoptimize x86 requantization by Marat Dukhan · 4 years, 2 months ago
  56. 2cc2622 Update links to TFLite/PyTorch integration in the README by Marat Dukhan · 4 years, 2 months ago
  57. d65a152 AVX2 versions of QS8 DWCONV microkernels by Marat Dukhan · 4 years, 2 months ago
  58. 0743cdf Hardcoded end-to-end benchmark on MobileNet v1 in QS8 format by Marat Dukhan · 4 years, 2 months ago
  59. 16f1e1a QS8 version of NHWC Convolution operator by Marat Dukhan · 4 years, 2 months ago
  60. 548e830 Fix NCW Global Average Pooling on WAsm SIMD by Marat Dukhan · 4 years, 2 months ago
  61. f62bbdc SSE2/SSSE3/SSE4.1/XOP implementation of QS8 DWCONV microkernels by Marat Dukhan · 4 years, 2 months ago
  62. 40bbafe NEON variants of QS8 GEMM & IGEMM microkernels by Marat Dukhan · 4 years, 2 months ago
  63. 22de5e7 WAsm SIMD implementation of QU8 Q31 requantization by Marat Dukhan · 4 years, 2 months ago
  64. ecd8311 Rename s8rng/s32rng -> i8rng/i32rng by Marat Dukhan · 4 years, 2 months ago
  65. 8ecf28d Update instruction count in Q31 QS8 WAsm SIMD requantization by Marat Dukhan · 4 years, 2 months ago
  66. dfcaa3c Include scalar QU8 requantization in the benchmark by Marat Dukhan · 4 years, 2 months ago
  67. 3e07b91 Reoptimize QS8 implementatation of Q31 requantization schema in WAsm SIMD by Marat Dukhan · 4 years, 2 months ago
  68. 56bdd4a QS8 requantization benchmark by Marat Dukhan · 4 years, 2 months ago
  69. 683fab3 XW (eXtended Weights) optimization for QS8 GEMM microkernel by Marat Dukhan · 4 years, 2 months ago
  70. 138560c WAsm SIMD version of QS8 requantization stub by Marat Dukhan · 4 years, 2 months ago
  71. e7edc80 Add 3x4c8 variants of SSE2/SSSE3/SSE4.1/XOP GEMM/IGEMM microkernels by Marat Dukhan · 4 years, 2 months ago
  72. 1280952 AVX2 version of QS8 GEMM and IGEMM microkernels by Marat Dukhan · 4 years, 2 months ago
  73. 1566fee XOP versions of QS8 GEMM/IGEMM microkernels by Marat Dukhan · 4 years, 2 months ago
  74. 44f0ca7 Bind RNG by reference in microbenchmarks by Marat Dukhan · 4 years, 2 months ago
  75. b02e522 Fix CMake build by Marat Dukhan · 4 years, 2 months ago
  76. 07bd252 QS8 IGEMM MRx4c8 SSE2/SSSE3/SSE4.1 microkernels by Marat Dukhan · 4 years, 2 months ago
  77. dee732b LD128 versions of QS8 GEMM SSE2/SSSE3/SSE4.1 microkernels by Marat Dukhan · 4 years, 2 months ago
  78. 14d3ce8 Add LD64 suffix in QS8 GEMM/IGEMM microkernels by Marat Dukhan · 4 years, 2 months ago
  79. 733d0be QS8 GEMM MRx4c8 SSE2/SSSE3/SSE4.1 microkernels by Marat Dukhan · 4 years, 2 months ago
  80. f948068 QS8 IGEMM microkernels and infrastructure by Marat Dukhan · 4 years, 2 months ago
  81. 595e170 QS8 GEMM microkernels and infrastructure by Marat Dukhan · 4 years, 2 months ago
  82. b7615dd Remove unused input_zero_point parameter for SSE2 QU8 GEMM by Marat Dukhan · 4 years, 2 months ago
  83. ec88e27 Refactor QU8 requantization params by Marat Dukhan · 4 years, 2 months ago
  84. 312b222 Remove unused requantization structures by Marat Dukhan · 4 years, 2 months ago
  85. 1065af4 Fix mismatch in parameter names in QU8 GEMM by Benoit Jacob · 4 years, 2 months ago
  86. 9199246 Remove redundant input_zero_point parameter from xnn_qu8_gemm_params by Marat Dukhan · 4 years, 2 months ago
  87. 2e23d2b Signed requantization evaluation stubs and unit tests by Marat Dukhan · 4 years, 2 months ago
  88. c5045bf Remove PSIMD variant of GAVGPOOL CW microkernel by Marat Dukhan · 4 years, 2 months ago
  89. e740cd5 Remove duplicated xnn_qu8_requantize_fp32__scalar_magic by Marat Dukhan · 4 years, 2 months ago
  90. 5b69f8b Rename requantization functions and filenames by Marat Dukhan · 4 years, 2 months ago
  91. 9531e9f Suffix VMULCADDC microkernels with activation name by Marat Dukhan · 4 years, 2 months ago
  92. a199d49 Remove support for direct Asm.js builds by Marat Dukhan · 4 years, 2 months ago
  93. 1849351 WAsm SIMD versions of GAVGPOOL CW microkernel by Marat Dukhan · 4 years, 2 months ago
  94. fc73f86 Remove PSIMD versions of ARGMAXPOOL microkernels by Marat Dukhan · 4 years, 2 months ago
  95. 92162da Remove PSIMD versions of PPMM and PACKX microkernels by Marat Dukhan · 4 years, 2 months ago
  96. ef25c6d NEON versions of ARGMAXPOOL microkernels by Marat Dukhan · 4 years, 2 months ago
  97. 3419206 Cleanup remntants of removed PSIMD microkernels by Marat Dukhan · 4 years, 2 months ago
  98. 0d0d882 WAsm SIMD version of PPMM microkernel by Marat Dukhan · 4 years, 2 months ago
  99. 1530116 Use more descriptive names in SpMM microkernels by Marat Dukhan · 4 years, 2 months ago
  100. b293e8d Fuse Clamp into Deconvolution/Divide/Subtract operators by Marat Dukhan · 4 years, 2 months ago