1. b0b66a7 Kelvin Resize nearest neighbors for int8 by Naveen Dodda · 2 weeks ago master
  2. 321c904 Rewrite ConvPerChannelD1OD24_5x5 by David Gao · 7 days ago
  3. 0758fb8 Improve Conv2dD32 by Alex Van Damme · 10 days ago
  4. 008f0ae Restrict ConvS16B64K1xn to height 1 by Alex Van Damme · 6 days ago
  5. f22bdde Fix in_channel calculation in ConvS8D4 by Alex Van Damme · 7 days ago
  6. 108cf32 Remove extra offset addition from ElementwiseAdd by Alex Van Damme · 7 days ago
  7. 5190b46 Merge "Add test_status calls to benchmark_smc" by Alex Van Damme · 11 days ago
  8. 35a67de Merge "Create templated version of Swizzle." by Alex Van Damme · 11 days ago
  9. 109726f Add test_status calls to benchmark_smc by Alex Van Damme · 11 days ago
  10. 5496087 Merge "Implement DepthwiseConvS85x5D32 w/ adwconv" by Alex Van Damme · 11 days ago
  11. 18694e4 Merge "Two outputs per iteration in ConvPerChannelD1OD24_5x5" by Alex Van Damme · 11 days ago
  12. 2cebe31 Implement DepthwiseConvS85x5D32 w/ adwconv by Alex Van Damme · 12 days ago
  13. f427e85 Create templated version of Swizzle. by Derek Chow · 12 days ago
  14. 41f1cec Reduce register pressure in ConvS8W8D4. by Derek Chow · 13 days ago
  15. f353919 Two outputs per iteration in ConvPerChannelD1OD24_5x5 by Alex Van Damme · 2 weeks ago
  16. 9e0b2ba Merge "Reduce vld calls in ConvS8D4." by Derek Chow · 13 days ago
  17. 2f60851 Reduce vld calls in ConvS8D4. by Derek Chow · 2 weeks ago
  18. 87cc267 Merge "Process 16 channels per iteration in 1x1 d16" by Alex Van Damme · 13 days ago
  19. cf069b3 Process 16 channels per iteration in 1x1 d16 by Alex Van Damme · 2 weeks ago
  20. b5caeab Merge "Improve ElementwiseAddS8" by Alex Van Damme · 2 weeks ago
  21. 74f2f74 Create and apply interleaved output pipeline macros by Alex Van Damme · 2 weeks ago
  22. 8da6e53 5x5 DepthwiseConv, 4 outputs per loop iteration by Alex Van Damme · 3 weeks ago
  23. ac91807 Improve ElementwiseAddS8 by Alex Van Damme · 3 weeks ago
  24. d22d7d6 DepthwiseConv 3x3, 4 per iteration by Alex Van Damme · 3 weeks ago
  25. 13fe02a Create RUN_KERNEL macro for DepthwiseConvS8 by Alex Van Damme · 3 weeks ago
  26. 088841b Specialize 3x3, stride 1 DepthwiseConv by Alex Van Damme · 3 weeks ago
  27. 5106685 Vectorized Logistic for int8 by Alex Van Damme · 4 weeks ago
  28. b309494 Merge changes I1f0ec9d3,I55e0cf22 by Cindy Liu · 3 weeks ago
  29. db10d3b Merge changes from topic "nexus-benchmark" by Cindy Liu · 3 weeks ago
  30. 627a3ba Lint cleanup benchmark_kelvin.cc by Cindy Liu · 3 weeks ago
  31. 2ec00d9 Fix carry over benchmark mismatch count by Cindy Liu · 3 weeks ago
  32. 8ef6f11 Use 1x1 d=32n kernel more aggressively. by Derek Chow · 3 weeks ago
  33. aaf9d30 Update benchmark device rule to support tags by Cindy Liu · 3 weeks ago
  34. f6ec59d Format benchmarks.bzl by Cindy Liu · 3 weeks ago
  35. 1ad0284 Point wise convolution input_depth = 16 by Derek Chow · 3 weeks ago
  36. 5160bc6 Add specialized pointwise convolution. by Derek Chow · 4 weeks ago
  37. dd44caa Allow specifying benchmark arena size by Alex Van Damme · 3 weeks ago
  38. d2aca97 Add additional ops to benchmark by Alex Van Damme · 4 weeks ago
  39. 78538b3 Use dwconv on input_depth == 1, 5x5 by Alex Van Damme · 5 weeks ago
  40. 0d75a42 Refactor kernel selection. by Derek Chow · 4 weeks ago
  41. 2aeaaa8 Enable O2 optimizations. by Derek Chow · 6 weeks ago
  42. 5971487 Improvements to elementwise_add_{s8,s16} by Alex Van Damme · 6 weeks ago
  43. ca72b40 Improve ConvS8D4 by Alex Van Damme · 6 weeks ago
  44. cd3d0e3 3x3 DepthwiseConv w/ adwconv by Alex Van Damme · 6 weeks ago
  45. b1afda6 Specialize 5x5 DepthwiseConv by Alex Van Damme · 7 weeks ago
  46. a99c93d Add support for checking output in benchmark by Alex Van Damme · 8 weeks ago
  47. fd54c9e Relax output_depth for ConvS8D4 by Alex Van Damme · 7 weeks ago
  48. 86806b3 Add input_depth == 1 specialized convolution. by Derek Chow · 7 weeks ago
  49. 456ed80 Merge "Support running a different Kelvin benchmark binary" by Michael Brooks · 6 weeks ago
  50. 40a8300 Fix swizzling and rounding for depthwise conv by Alex Van Damme · 7 weeks ago
  51. e617f61 Relax restrictions on aconv-based Conv2D by Alex Van Damme · 7 weeks ago
  52. 3b8d3cb sw/kelvin: Refactoring TFLM Kelvin kernel optimization codes by Lun Dong · 7 weeks ago
  53. d7880e8 Add doc on tflm op support in Kelvin. by Derek Chow · 7 weeks ago
  54. 2d626f6 Support running a different Kelvin benchmark binary by Michael Brooks · 7 weeks ago
  55. b11ad17 Specialized int8 conv kernels by Alex Van Damme · 10 weeks ago
  56. be4ab97 WIP Depthwise conv for 2d kernels by Naveen Dodda · 10 weeks ago
  57. c62ec01 Add wall-time measurement to Benchmark by Alex Van Damme · 3 months ago
  58. c47ddc8 Increasing memory to support models for benchmarks. by Naveen Dodda · 3 months ago
  59. 1eb552e Updating resolver to support mobilenet models by Naveen Dodda · 3 months ago
  60. ff78446 Add support for using test data in model benchmarks by Michael Brooks · 3 months ago
  61. 5f67634 Merge "Add Support for Profiling TFLM Models in Benchmarks" by Michael Brooks · 3 months ago
  62. d884926 Fix hw test script to support extra flags by Cindy Liu · 3 months ago
  63. fe2b2a8 Add Support for Profiling TFLM Models in Benchmarks by Michael Brooks · 3 months ago
  64. a1e166f Pass through extra flags to the hardware sim by Julian Mullings-Black · 3 months ago
  65. d48c5a4 Patches for benchmark on ASIC by Alex Van Damme · 3 months ago
  66. 3bdf8f4 Add minstret to perf_counters by Alex Van Damme · 5 months ago
  67. c0214c3 Add ecall/mret test by Alex Van Damme · 4 months ago
  68. 8d688e2 Fix Benchmark SMC Print Format by Michael Brooks · 3 months ago
  69. 03b3adf Additonal Benchmark Targets by Michael Brooks · 3 months ago
  70. dc03ded Add benchmark suite to sw/kelvin by Michael Brooks · 3 months ago
  71. 215d48a Enabling kelvin_test outside of sw/kelvin. by Naveen Dodda · 5 months ago
  72. 0b2b7fb Type-check around find_cc_toolchain result usage by Alex Van Damme · 5 months ago
  73. 2b6d459 Mark `model_input` and `output_header` as NOLOAD by Alex Van Damme · 7 months ago
  74. 0cc464f Make pooling_test "large" to adjust timeout by Julian Mullings-Black · 5 months ago
  75. 24619c5 Add perf_counters test by Alex Van Damme · 6 months ago
  76. 78e3bc3 Update .bazelversion to 6.2.1 by Alex Van Damme · 7 months ago
  77. 3285c6a Changes to optimize max_pool for Kelvin int8 dtype by Naveen-Dodda · 7 months ago
  78. 0c9118e Add riscv CSR tests, enable zicsr by Alex Van Damme · 7 months ago
  79. 4054485 Merge "sw/kelvin: revamp layout to support setting model input data" by Sam Leffler · 7 months ago
  80. c231ced sw/kelvin: revamp layout to support setting model input data by Sam Leffler · 8 months ago
  81. 903c311 Consolidate kernel selection behind generic conv_per_channel_b64. by Derek Chow · 7 months ago
  82. a5f129c Optimize 1x1 group conv. by Derek Chow · 8 months ago
  83. 385ea39 Fix PrintfTraits print out by Cindy Liu · 8 months ago
  84. 2095c66 Update vsran unsigned tests by Cindy Liu · 8 months ago
  85. 5a2cf62 Add vst unit test by Cindy Liu · 8 months ago
  86. 9c9167a Cleanup kelvin_isa test format warnings by Cindy Liu · 8 months ago
  87. ac27a72 Add vstq unit test by Cindy Liu · 8 months ago
  88. 76d4adc Add vsrn and vzip unit tests by Cindy Liu · 8 months ago
  89. 503e5ce Add more vector unit tests by Cindy Liu · 8 months ago
  90. 2fe3247 Add acset actr unit tests by Cindy Liu · 8 months ago
  91. 6448eab Add unit tests by Cindy Liu · 8 months ago
  92. 3ffc473 OSS cleanup on the host_tool by Cindy Liu · 8 months ago
  93. 66f7e3c Fix contributing doc code review description by Cindy Liu · 8 months ago
  94. 43879e4 OSS cleanup by Cindy Liu · 8 months ago
  95. c76bc0c Add depthwise conv test. by Derek Chow · 9 months ago
  96. 766e5af Add dilation support for depthwise convolution. by Derek Chow · 9 months ago
  97. 16ffb60 8x16 Depthwise convolution for Kelvin. by Derek Chow · 9 months ago
  98. f39cadd Helper methods for Conv2D kernel by Alex Van Damme · 10 months ago
  99. fc5baaa Move soundstream TFLM example to the model repository by Cindy Liu · 9 months ago
  100. e525be0 Remove remaining C++ from kelvin.h, include in hello_world by Alex Van Damme · 9 months ago