Created
May 15, 2023 18:19
-
-
Save HDCharles/c7d2b88475a2695a12d4eedf3c7f0ce3 to your computer and use it in GitHub Desktop.
benchmark_safe_int_mm.py
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
shape_x shape_w mm_half_ms int_mm_ms safe_int_mm_ms trit_int_mm_ms trit_int_t_mm_ms int_mm_speedup safe_int_mm_speedup int_mm_trit_speedup int_mm_trit_t_speedup | |
-------------- -------------- ------------ ----------- ---------------- ---------------- ------------------ ---------------- --------------------- --------------------- ----------------------- | |
(512, 512) (512, 512) 13.9634 30.6949 30.7042 59.4251 59.263 0.454909 0.454773 0.234975 0.235618 | |
(512, 512) (512, 2048) 18.0451 30.7888 30.7703 59.1577 59.7134 0.586093 0.586445 0.305034 0.302195 | |
(512, 512) (512, 16384) 48.1365 130.105 130.148 174.705 78.7401 0.369981 0.369859 0.275531 0.611334 | |
(512, 2048) (2048, 512) 13.8353 101.849 101.859 76.8298 59.6942 0.13584 0.135828 0.180077 0.231769 | |
(512, 2048) (2048, 2048) 30.9815 102.168 102.175 149.542 59.8102 0.30324 0.303221 0.207176 0.517998 | |
(512, 2048) (2048, 16384) 154.327 487.136 487.188 529.619 172.787 0.316806 0.316772 0.291393 0.893164 | |
(512, 16384) (16384, 512) 54.02 767.547 767.521 592.561 134.169 0.0703801 0.0703825 0.0911636 0.402625 | |
(512, 16384) (16384, 2048) 159.409 768.655 768.781 1174.75 249.569 0.207387 0.207353 0.135696 0.638739 | |
(512, 16384) (16384, 16384) 1102.17 3809.39 3809.41 3842.43 1052.77 0.28933 0.289328 0.286842 1.04693 | |
(2048, 512) (512, 512) 17.6008 30.7604 30.7606 59.7059 60.141 0.572189 0.572187 0.294792 0.292659 | |
(2048, 512) (512, 2048) 24.3038 80.4719 80.4706 105.475 59.9677 0.302016 0.302021 0.230422 0.405281 | |
(2048, 512) (512, 16384) 167.65 475.945 475.812 495.035 176.815 0.352246 0.352344 0.338662 0.948162 | |
(2048, 2048) (2048, 512) 31.2354 102.152 102.224 148.881 59.908 0.305774 0.305558 0.209801 0.521389 | |
(2048, 2048) (2048, 2048) 83.0828 294.598 294.297 327.471 95.6765 0.282021 0.282309 0.25371 0.868371 | |
(2048, 2048) (2048, 16384) 576.986 1832.74 1829.84 1580.82 408.565 0.314821 0.31532 0.364991 1.41223 | |
(2048, 16384) (16384, 512) 168.067 769.117 769.109 1112.09 239.809 0.218519 0.218521 0.151126 0.700835 | |
(2048, 16384) (16384, 2048) 577.218 2293.53 2290.08 2391.11 526.582 0.251673 0.252051 0.241402 1.09616 | |
(2048, 16384) (16384, 16384) 4287.56 14476.1 14458.8 11855.5 2565.82 0.296182 0.296536 0.361651 1.67103 | |
(16384, 512) (512, 512) 47.7314 130.118 130.138 172.904 78.7159 0.366831 0.366776 0.276057 0.606375 | |
(16384, 512) (512, 2048) 165.87 476.028 476.161 494.383 176.355 0.348446 0.348349 0.33551 0.94055 | |
(16384, 512) (512, 16384) 1268.18 3754.19 3754.22 3500.15 1084.49 0.337805 0.337802 0.362322 1.16938 | |
(16384, 2048) (2048, 512) 167.052 487.932 487.907 528.668 171.443 0.342366 0.342384 0.315986 0.974385 | |
(16384, 2048) (2048, 2048) 568.951 1829.75 1830.06 1587.88 408.961 0.310944 0.310892 0.358309 1.39121 | |
(16384, 2048) (2048, 16384) 4756.14 14603.3 14577.5 11505.3 2642.91 0.32569 0.326266 0.413387 1.79959 | |
(16384, 16384) (16384, 512) 1101.84 3812.3 3812.13 3849.53 1049.88 0.289024 0.289036 0.286228 1.0495 | |
(16384, 16384) (16384, 2048) 4182.75 14468.3 14711.1 11864.8 2564.23 0.289098 0.284325 0.352534 1.63119 | |
(16384, 16384) (16384, 16384) 33442.9 115603 115624 87274.2 17419.2 0.28929 0.289239 0.383193 1.91989 |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment