Skip to content
This repository has been archived by the owner on Sep 25, 2023. It is now read-only.

[REVIEW] Various optimization across all functions #267

Merged
merged 59 commits into from
Oct 16, 2020
Merged

[REVIEW] Various optimization across all functions #267

merged 59 commits into from
Oct 16, 2020

Conversation

mnicely
Copy link
Contributor

@mnicely mnicely commented Oct 11, 2020 •
edited
Loading

The PR:

  1. Implement a style standard of add C++11 to all Elementwise kernels.
  2. Replace cupy.fftpack with cupy.fft
  3. Series of elementwise kernels optimizations.
  4. Clean up logic to call CUDA kernels.
  5. Move firwin to CPU, expected data size too small (~2x resample_poly)

@mnicely mnicely added the 2 - In Progress Currenty a work in progress label Oct 11, 2020
@mnicely mnicely requested a review from awthomp October 11, 2020 18:32
@mnicely mnicely requested a review from a team as a code owner October 11, 2020 18:32
@mnicely mnicely self-assigned this Oct 11, 2020
@GPUtester
Copy link
Contributor

Please update the changelog in order to start CI tests.

View the gpuCI docs here.

@awthomp
Copy link
Member

awthomp commented Oct 11, 2020

Hey @mnicely, can you change the title of this PR to account for the fftpack changes too?

@mnicely mnicely changed the base branch from branch-0.16 to branch-0.17 October 12, 2020 12:57
@mnicely mnicely changed the title [WIP] Add c++11 to Elementwise Kernel [WIP] Various optimization across all functions Oct 12, 2020
@BradReesWork BradReesWork added this to the 0.17 milestone Oct 14, 2020
@mnicely
Copy link
Contributor Author

mnicely commented Oct 15, 2020

PR comparison

Computing stats ...
--------------------- benchmark 'Argrelextrema': 24 tests ---------------------
Name (time in us)                                                Mean          
-------------------------------------------------------------------------------
test_argrelextrema_gpu[clip-1--1-1-32768] (0004_updates)      98.9318 (1.0)    
test_argrelextrema_gpu[clip-2--1-1-32768] (0004_updates)      99.9253 (1.01)   
test_argrelextrema_gpu[wrap-2--1-1-32768] (0004_updates)     100.1835 (1.01)   
test_argrelextrema_gpu[wrap-1--1-1-32768] (0004_updates)     100.2787 (1.01)   
test_argrelextrema_gpu[clip-2--1-1-32768] (0005_branch-)     101.0654 (1.02)   
test_argrelextrema_gpu[wrap-2--1-1-32768] (0005_branch-)     101.2414 (1.02)   
test_argrelextrema_gpu[wrap-1--1-1-32768] (0005_branch-)     101.5297 (1.03)   
test_argrelextrema_gpu[clip-1--1-1-32768] (0005_branch-)     103.2746 (1.04)   
test_argrelextrema_gpu[wrap-2--1-2-256] (0004_updates)       103.3428 (1.04)   
test_argrelextrema_gpu[clip-2--1-2-256] (0004_updates)       103.4321 (1.05)   
test_argrelextrema_gpu[clip-1--1-2-256] (0004_updates)       104.6964 (1.06)   
test_argrelextrema_gpu[wrap-1--1-2-256] (0004_updates)       105.4092 (1.07)   
test_argrelextrema_gpu[clip-1--1-2-256] (0005_branch-)       106.3931 (1.08)   
test_argrelextrema_gpu[wrap-1--1-2-256] (0005_branch-)       106.5218 (1.08)   
test_argrelextrema_gpu[wrap-2--1-2-256] (0005_branch-)       106.6009 (1.08)   
test_argrelextrema_gpu[clip-2--1-2-256] (0005_branch-)       107.0982 (1.08)   
test_argrelextrema_gpu[wrap-1--1-3-32] (0004_updates)        272.1056 (2.75)   
test_argrelextrema_gpu[wrap-1--1-3-32] (0005_branch-)        278.0810 (2.81)   
test_argrelextrema_gpu[clip-1--1-3-32] (0004_updates)        325.5380 (3.29)   
test_argrelextrema_gpu[clip-1--1-3-32] (0005_branch-)        330.2460 (3.34)   
test_argrelextrema_gpu[wrap-2--1-3-32] (0004_updates)        403.8022 (4.08)   
test_argrelextrema_gpu[wrap-2--1-3-32] (0005_branch-)        412.3947 (4.17)   
test_argrelextrema_gpu[clip-2--1-3-32] (0005_branch-)        505.8483 (5.11)   
test_argrelextrema_gpu[clip-2--1-3-32] (0004_updates)        520.6247 (5.26)   
-------------------------------------------------------------------------------


--------------------- benchmark 'Argrelmin': 24 tests ---------------------
Name (time in us)                                            Mean          
---------------------------------------------------------------------------
test_argrelmin_gpu[clip-2--1-1-32768] (0004_updates)     100.2999 (1.0)    
test_argrelmin_gpu[clip-1--1-1-32768] (0004_updates)     100.4905 (1.00)   
test_argrelmin_gpu[wrap-2--1-1-32768] (0004_updates)     100.6505 (1.00)   
test_argrelmin_gpu[wrap-1--1-1-32768] (0004_updates)     101.2342 (1.01)   
test_argrelmin_gpu[clip-1--1-1-32768] (0005_branch-)     103.8873 (1.04)   
test_argrelmin_gpu[clip-1--1-2-256] (0004_updates)       104.1420 (1.04)   
test_argrelmin_gpu[wrap-1--1-1-32768] (0005_branch-)     104.6143 (1.04)   
test_argrelmin_gpu[clip-2--1-2-256] (0004_updates)       104.7754 (1.04)   
test_argrelmin_gpu[wrap-2--1-1-32768] (0005_branch-)     104.8680 (1.05)   
test_argrelmin_gpu[wrap-1--1-2-256] (0004_updates)       105.0144 (1.05)   
test_argrelmin_gpu[wrap-2--1-2-256] (0004_updates)       105.4472 (1.05)   
test_argrelmin_gpu[clip-2--1-1-32768] (0005_branch-)     105.6745 (1.05)   
test_argrelmin_gpu[clip-1--1-2-256] (0005_branch-)       109.3220 (1.09)   
test_argrelmin_gpu[wrap-2--1-2-256] (0005_branch-)       109.5431 (1.09)   
test_argrelmin_gpu[clip-2--1-2-256] (0005_branch-)       110.2886 (1.10)   
test_argrelmin_gpu[wrap-1--1-2-256] (0005_branch-)       110.6544 (1.10)   
test_argrelmin_gpu[wrap-1--1-3-32] (0004_updates)        272.9932 (2.72)   
test_argrelmin_gpu[wrap-1--1-3-32] (0005_branch-)        286.1029 (2.85)   
test_argrelmin_gpu[clip-1--1-3-32] (0004_updates)        330.7360 (3.30)   
test_argrelmin_gpu[clip-1--1-3-32] (0005_branch-)        338.6316 (3.38)   
test_argrelmin_gpu[wrap-2--1-3-32] (0004_updates)        408.2672 (4.07)   
test_argrelmin_gpu[wrap-2--1-3-32] (0005_branch-)        417.3933 (4.16)   
test_argrelmin_gpu[clip-2--1-3-32] (0004_updates)        515.7716 (5.14)   
test_argrelmin_gpu[clip-2--1-3-32] (0005_branch-)        522.1726 (5.21)   
---------------------------------------------------------------------------


---------------- benchmark 'Barlett': 2 tests ---------------
Name (time in us)                              Mean          
-------------------------------------------------------------
test_bartlett_gpu[32768] (0004_updates)     13.2932 (1.0)    
test_bartlett_gpu[32768] (0005_branch-)     14.4236 (1.09)   
-------------------------------------------------------------


--------------- benchmark 'BartHann': 2 tests ---------------
Name (time in us)                              Mean          
-------------------------------------------------------------
test_barthann_gpu[32768] (0004_updates)     15.0997 (1.0)    
test_barthann_gpu[32768] (0005_branch-)     18.4420 (1.22)   
-------------------------------------------------------------


--------------- benchmark 'Blackman': 2 tests ---------------
Name (time in us)                              Mean          
-------------------------------------------------------------
test_blackman_gpu[32768] (0004_updates)     51.1566 (1.0)    
test_blackman_gpu[32768] (0005_branch-)     53.0598 (1.04)   
-------------------------------------------------------------


--------------- benchmark 'BlackmanHarris': 2 tests ---------------
Name (time in us)                                    Mean          
-------------------------------------------------------------------
test_blackmanharris_gpu[32768] (0004_updates)     59.2638 (1.0)    
test_blackmanharris_gpu[32768] (0005_branch-)     61.7481 (1.04)   
-------------------------------------------------------------------


--------------- benchmark 'Bohman': 2 tests ---------------
Name (time in us)                            Mean          
-----------------------------------------------------------
test_bohman_gpu[32768] (0004_updates)     17.8453 (1.0)    
test_bohman_gpu[32768] (0005_branch-)     21.2621 (1.19)   
-----------------------------------------------------------


--------------- benchmark 'Boxcar': 2 tests ---------------
Name (time in us)                            Mean          
-----------------------------------------------------------
test_boxcar_gpu[32768] (0004_updates)     13.0072 (1.0)    
test_boxcar_gpu[32768] (0005_branch-)     13.4079 (1.03)   
-----------------------------------------------------------


---------------------------- benchmark 'CSD': 16 tests ----------------------------
Name (time in us)                                                    Mean          
-----------------------------------------------------------------------------------
test_csd_gpu[2048-1.0-16384-float64] (0004_updates)              464.4981 (1.0)    
test_csd_gpu[1024-1000000.0-16384-float64] (0004_updates)        466.1994 (1.00)   
test_csd_gpu[2048-1000000.0-16384-float64] (0004_updates)        466.4274 (1.00)   
test_csd_gpu[1024-1.0-16384-float64] (0004_updates)              467.0230 (1.01)   
test_csd_gpu[2048-1.0-16384-float64] (0005_branch-)              472.6453 (1.02)   
test_csd_gpu[2048-1000000.0-16384-float64] (0005_branch-)        473.0760 (1.02)   
test_csd_gpu[1024-1.0-16384-float64] (0005_branch-)              474.1645 (1.02)   
test_csd_gpu[1024-1000000.0-16384-float64] (0005_branch-)        475.7479 (1.02)   
test_csd_gpu[1024-1000000.0-16384-complex128] (0004_updates)     495.0340 (1.07)   
test_csd_gpu[1024-1000000.0-16384-complex128] (0005_branch-)     495.7958 (1.07)   
test_csd_gpu[2048-1000000.0-16384-complex128] (0004_updates)     496.7682 (1.07)   
test_csd_gpu[2048-1000000.0-16384-complex128] (0005_branch-)     499.9129 (1.08)   
test_csd_gpu[1024-1.0-16384-complex128] (0005_branch-)           502.4013 (1.08)   
test_csd_gpu[1024-1.0-16384-complex128] (0004_updates)           502.7583 (1.08)   
test_csd_gpu[2048-1.0-16384-complex128] (0005_branch-)           505.7827 (1.09)   
test_csd_gpu[2048-1.0-16384-complex128] (0004_updates)           506.9083 (1.09)   
-----------------------------------------------------------------------------------


----------------------- benchmark 'CWT': 8 tests ----------------------
Name (time in ms)                                        Mean          
-----------------------------------------------------------------------
test_cwt_gpu[31-16384-complex128] (0005_branch-)      12.6480 (1.0)    
test_cwt_gpu[31-16384-float64] (0004_updates)         13.4957 (1.07)   
test_cwt_gpu[31-16384-complex128] (0004_updates)      13.8086 (1.09)   
test_cwt_gpu[31-16384-float64] (0005_branch-)         13.9956 (1.11)   
test_cwt_gpu[127-16384-complex128] (0004_updates)     51.6457 (4.08)   
test_cwt_gpu[127-16384-complex128] (0005_branch-)     54.1012 (4.28)   
test_cwt_gpu[127-16384-float64] (0004_updates)        55.7404 (4.41)   
test_cwt_gpu[127-16384-float64] (0005_branch-)        58.0863 (4.59)   
-----------------------------------------------------------------------


------------------------- benchmark 'ChannelizePoly': 24 tests ------------------------
Name (time in us)                                                        Mean          
---------------------------------------------------------------------------------------
test_channelizepoly_gpu[128-2048-4096-float32] (0005_branch-)         69.2687 (1.0)    
test_channelizepoly_gpu[64-2048-4096-float32] (0005_branch-)          69.2903 (1.00)   
test_channelizepoly_gpu[128-2048-4096-complex64] (0005_branch-)       69.7155 (1.01)   
test_channelizepoly_gpu[64-2048-4096-float64] (0005_branch-)          70.2351 (1.01)   
test_channelizepoly_gpu[128-2048-4096-float64] (0005_branch-)         70.3340 (1.02)   
test_channelizepoly_gpu[64-2048-4096-complex64] (0005_branch-)        70.3724 (1.02)   
test_channelizepoly_gpu[64-2048-4096-complex64] (0004_updates)        70.6417 (1.02)   
test_channelizepoly_gpu[128-2048-4096-complex128] (0005_branch-)      70.7681 (1.02)   
test_channelizepoly_gpu[64-2048-4096-float32] (0004_updates)          70.8618 (1.02)   
test_channelizepoly_gpu[128-2048-4096-complex64] (0004_updates)       71.1981 (1.03)   
test_channelizepoly_gpu[64-2048-4096-complex128] (0005_branch-)       71.2370 (1.03)   
test_channelizepoly_gpu[128-2048-4096-float32] (0004_updates)         71.2401 (1.03)   
test_channelizepoly_gpu[128-2048-4096-complex128] (0004_updates)      71.9167 (1.04)   
test_channelizepoly_gpu[64-2048-4096-float64] (0004_updates)          72.0412 (1.04)   
test_channelizepoly_gpu[64-2048-4096-complex128] (0004_updates)       72.4387 (1.05)   
test_channelizepoly_gpu[128-2048-4096-float64] (0004_updates)         73.2789 (1.06)   
test_channelizepoly_gpu[256-2048-4096-float32] (0004_updates)         81.7889 (1.18)   
test_channelizepoly_gpu[256-2048-4096-complex64] (0004_updates)       82.5843 (1.19)   
test_channelizepoly_gpu[256-2048-4096-float32] (0005_branch-)         83.2703 (1.20)   
test_channelizepoly_gpu[256-2048-4096-complex64] (0005_branch-)       83.8634 (1.21)   
test_channelizepoly_gpu[256-2048-4096-float64] (0004_updates)         86.7559 (1.25)   
test_channelizepoly_gpu[256-2048-4096-float64] (0005_branch-)         89.0475 (1.29)   
test_channelizepoly_gpu[256-2048-4096-complex128] (0004_updates)     134.8547 (1.95)   
test_channelizepoly_gpu[256-2048-4096-complex128] (0005_branch-)     136.9829 (1.98)   
---------------------------------------------------------------------------------------


------------------ benchmark 'Chebwin': 6 tests -----------------
Name (time in us)                                  Mean          
-----------------------------------------------------------------
test_chebwin_gpu[100-32768] (0004_updates)     126.7690 (1.0)    
test_chebwin_gpu[50-32768] (0004_updates)      127.2702 (1.00)   
test_chebwin_gpu[100-32768] (0005_branch-)     134.9484 (1.06)   
test_chebwin_gpu[50-32768] (0005_branch-)      135.1321 (1.07)   
test_chebwin_gpu[50-32767] (0004_updates)      162.4243 (1.28)   
test_chebwin_gpu[100-32767] (0004_updates)     162.7607 (1.28)   
-----------------------------------------------------------------


--------------------- benchmark 'Chirp': 8 tests ---------------------
Name (time in us)                                       Mean          
----------------------------------------------------------------------
test_chirp_gpu[lin-10-1-6-16384] (0004_updates)      60.7337 (1.0)    
test_chirp_gpu[quad-10-1-6-16384] (0004_updates)     61.4509 (1.01)   
test_chirp_gpu[quad-10-1-6-16384] (0005_branch-)     63.1975 (1.04)   
test_chirp_gpu[lin-10-1-6-16384] (0005_branch-)      63.2041 (1.04)   
test_chirp_gpu[hyp-10-1-6-16384] (0004_updates)      64.2599 (1.06)   
test_chirp_gpu[hyp-10-1-6-16384] (0005_branch-)      66.2691 (1.09)   
test_chirp_gpu[log-10-1-6-16384] (0004_updates)      70.9446 (1.17)   
test_chirp_gpu[log-10-1-6-16384] (0005_branch-)      73.3140 (1.21)   
----------------------------------------------------------------------


-------------- benchmark 'CmplxSort': 10 tests --------------
Name (time in us)                              Mean          
-------------------------------------------------------------
test_cmplx_sort_gpu[p0] (0004_updates)     120.8413 (1.0)    
test_cmplx_sort_gpu[p2] (0004_updates)     121.4299 (1.00)   
test_cmplx_sort_gpu[p1] (0004_updates)     121.4597 (1.01)   
test_cmplx_sort_gpu[p3] (0004_updates)     121.6628 (1.01)   
test_cmplx_sort_gpu[p4] (0004_updates)     121.9892 (1.01)   
test_cmplx_sort_gpu[p4] (0005_branch-)     123.1248 (1.02)   
test_cmplx_sort_gpu[p0] (0005_branch-)     123.2948 (1.02)   
test_cmplx_sort_gpu[p3] (0005_branch-)     123.3308 (1.02)   
test_cmplx_sort_gpu[p2] (0005_branch-)     123.5205 (1.02)   
test_cmplx_sort_gpu[p1] (0005_branch-)     123.8755 (1.03)   
-------------------------------------------------------------


--------------------------- benchmark 'Coherence': 16 tests ---------------------------
Name (time in ms)                                                        Mean          
---------------------------------------------------------------------------------------
test_coherence_gpu[1024-1000000.0-16384-float64] (0004_updates)        1.2684 (1.0)    
test_coherence_gpu[2048-1.0-16384-float64] (0004_updates)              1.2700 (1.00)   
test_coherence_gpu[2048-1000000.0-16384-float64] (0004_updates)        1.2709 (1.00)   
test_coherence_gpu[2048-1.0-16384-float64] (0005_branch-)              1.2713 (1.00)   
test_coherence_gpu[1024-1000000.0-16384-float64] (0005_branch-)        1.2753 (1.01)   
test_coherence_gpu[2048-1000000.0-16384-float64] (0005_branch-)        1.2787 (1.01)   
test_coherence_gpu[1024-1.0-16384-float64] (0004_updates)              1.2791 (1.01)   
test_coherence_gpu[1024-1.0-16384-float64] (0005_branch-)              1.2978 (1.02)   
test_coherence_gpu[1024-1000000.0-16384-complex128] (0004_updates)     1.3343 (1.05)   
test_coherence_gpu[2048-1.0-16384-complex128] (0004_updates)           1.3462 (1.06)   
test_coherence_gpu[1024-1.0-16384-complex128] (0004_updates)           1.3493 (1.06)   
test_coherence_gpu[1024-1.0-16384-complex128] (0005_branch-)           1.3603 (1.07)   
test_coherence_gpu[1024-1000000.0-16384-complex128] (0005_branch-)     1.3618 (1.07)   
test_coherence_gpu[2048-1.0-16384-complex128] (0005_branch-)           1.3631 (1.07)   
test_coherence_gpu[2048-1000000.0-16384-complex128] (0005_branch-)     1.3639 (1.08)   
test_coherence_gpu[2048-1000000.0-16384-complex128] (0004_updates)     1.3834 (1.09)   
---------------------------------------------------------------------------------------


------------------ benchmark 'ComplexCepstrum': 8 tests ------------------
Name (time in us)                                           Mean          
--------------------------------------------------------------------------
test_complex_cepstrum_gpu[256-256] (0004_updates)       202.8165 (1.0)    
test_complex_cepstrum_gpu[256-16384] (0004_updates)     222.0658 (1.09)   
test_complex_cepstrum_gpu[123-16384] (0004_updates)     223.0617 (1.10)   
test_complex_cepstrum_gpu[123-256] (0004_updates)       226.9838 (1.12)   
test_complex_cepstrum_gpu[123-256] (0005_branch-)       395.9248 (1.95)   
test_complex_cepstrum_gpu[256-256] (0005_branch-)       396.2293 (1.95)   
test_complex_cepstrum_gpu[123-16384] (0005_branch-)     398.4711 (1.96)   
test_complex_cepstrum_gpu[256-16384] (0005_branch-)     404.0844 (1.99)   
--------------------------------------------------------------------------


------------------------- benchmark 'Convolve': 162 tests -------------------------
Name (time in us)                                                    Mean          
-----------------------------------------------------------------------------------
test_convolve1d_gpu[direct-same-256-128] (0004_updates)          103.0020 (1.0)    
test_convolve1d_gpu[direct-same-8192-128] (0004_updates)         103.1829 (1.00)   
test_convolve1d_gpu[direct-same-125-128] (0004_updates)          103.4874 (1.00)   
test_convolve1d_gpu[direct-same-8192-128] (0005_branch-)         103.7340 (1.01)   
test_convolve1d_gpu[direct-full-125-128] (0004_updates)          105.0298 (1.02)   
test_convolve1d_gpu[direct-full-125-128] (0005_branch-)          105.4635 (1.02)   
test_convolve1d_gpu[direct-same-125-128] (0005_branch-)          105.5532 (1.02)   
test_convolve1d_gpu[direct-valid-125-128] (0004_updates)         106.1959 (1.03)   
test_convolve1d_gpu[direct-valid-125-128] (0005_branch-)         106.2048 (1.03)   
test_convolve1d_gpu[direct-valid-256-128] (0004_updates)         106.5615 (1.03)   
test_convolve1d_gpu[direct-valid-256-128] (0005_branch-)         106.7884 (1.04)   
test_convolve1d_gpu[direct-same-256-128] (0005_branch-)          106.9587 (1.04)   
test_convolve1d_gpu[direct-full-256-128] (0005_branch-)          116.3844 (1.13)   
test_convolve1d_gpu[direct-full-256-128] (0004_updates)          117.2074 (1.14)   
test_convolve1d_gpu[auto-same-125-128] (0004_updates)            118.1194 (1.15)   
test_convolve1d_gpu[auto-same-125-128] (0005_branch-)            121.6631 (1.18)   
test_convolve1d_gpu[auto-full-125-128] (0005_branch-)            121.8308 (1.18)   
test_convolve1d_gpu[auto-valid-256-128] (0004_updates)           122.5433 (1.19)   
test_convolve1d_gpu[auto-full-125-128] (0004_updates)            123.8686 (1.20)   
test_convolve1d_gpu[auto-valid-125-128] (0005_branch-)           124.0887 (1.20)   
test_convolve1d_gpu[auto-valid-125-128] (0004_updates)           125.0846 (1.21)   
test_convolve1d_gpu[auto-valid-256-128] (0005_branch-)           127.3249 (1.24)   
test_convolve1d_gpu[direct-same-125-1025] (0004_updates)         131.6982 (1.28)   
test_convolve1d_gpu[direct-full-8192-128] (0004_updates)         131.7760 (1.28)   
test_convolve1d_gpu[direct-full-8192-128] (0005_branch-)         131.7797 (1.28)   
test_convolve1d_gpu[direct-valid-125-1025] (0004_updates)        132.2725 (1.28)   
test_convolve1d_gpu[direct-same-125-1025] (0005_branch-)         132.7369 (1.29)   
test_convolve1d_gpu[auto-full-256-128] (0005_branch-)            132.8953 (1.29)   
test_convolve1d_gpu[direct-valid-125-1025] (0005_branch-)        133.5855 (1.30)   
test_convolve1d_gpu[auto-full-256-128] (0004_updates)            134.7634 (1.31)   
test_convolve1d_gpu[direct-valid-8192-128] (0005_branch-)        135.7612 (1.32)   
test_convolve1d_gpu[direct-valid-8192-128] (0004_updates)        136.6712 (1.33)   
test_convolve1d_gpu[direct-same-125-8192] (0004_updates)         139.2633 (1.35)   
test_convolve1d_gpu[direct-same-125-8192] (0005_branch-)         140.3924 (1.36)   
test_convolve1d_gpu[direct-full-125-8192] (0004_updates)         143.8696 (1.40)   
test_convolve1d_gpu[direct-full-125-1025] (0004_updates)         144.2505 (1.40)   
test_convolve1d_gpu[direct-valid-125-8192] (0004_updates)        144.9292 (1.41)   
test_convolve1d_gpu[direct-valid-125-8192] (0005_branch-)        145.0674 (1.41)   
test_convolve1d_gpu[direct-full-125-8192] (0005_branch-)         145.2054 (1.41)   
test_convolve1d_gpu[direct-full-125-1025] (0005_branch-)         146.2952 (1.42)   
test_convolve1d_gpu[auto-valid-125-1025] (0005_branch-)          155.9061 (1.51)   
test_convolve1d_gpu[auto-valid-125-1025] (0004_updates)          161.5749 (1.57)   
test_convolve1d_gpu[direct-valid-256-1025] (0004_updates)        164.5469 (1.60)   
test_convolve1d_gpu[direct-valid-256-1025] (0005_branch-)        165.1128 (1.60)   
test_convolve1d_gpu[auto-full-125-1025] (0005_branch-)           165.1155 (1.60)   
test_convolve1d_gpu[direct-same-256-8192] (0004_updates)         165.3770 (1.61)   
test_convolve1d_gpu[auto-full-125-1025] (0004_updates)           165.7222 (1.61)   
test_convolve1d_gpu[direct-same-256-8192] (0005_branch-)         168.1799 (1.63)   
test_convolve1d_gpu[direct-full-256-8192] (0004_updates)         168.7423 (1.64)   
test_convolve1d_gpu[direct-full-256-8192] (0005_branch-)         169.4697 (1.65)   
test_convolve1d_gpu[direct-valid-256-8192] (0004_updates)        169.8632 (1.65)   
test_convolve1d_gpu[direct-valid-256-8192] (0005_branch-)        171.5115 (1.67)   
test_convolve1d_gpu[direct-same-256-1025] (0004_updates)         174.2512 (1.69)   
test_convolve1d_gpu[direct-full-256-1025] (0005_branch-)         175.1315 (1.70)   
test_convolve1d_gpu[direct-full-256-1025] (0004_updates)         175.9305 (1.71)   
test_convolve1d_gpu[direct-same-256-1025] (0005_branch-)         176.1235 (1.71)   
test_convolve1d_gpu[auto-valid-256-1025] (0005_branch-)          182.4475 (1.77)   
test_convolve1d_gpu[auto-valid-256-1025] (0004_updates)          187.7957 (1.82)   
test_convolve1d_gpu[fft-full-125-128] (0005_branch-)             333.4442 (3.24)   
test_convolve1d_gpu[fft-full-256-128] (0005_branch-)             335.8015 (3.26)   
test_convolve1d_gpu[fft-full-256-128] (0004_updates)             341.9479 (3.32)   
test_convolve1d_gpu[fft-full-125-128] (0004_updates)             344.3611 (3.34)   
test_convolve1d_gpu[fft-same-125-128] (0005_branch-)             346.8480 (3.37)   
test_convolve1d_gpu[fft-same-125-128] (0004_updates)             347.5164 (3.37)   
test_convolve1d_gpu[fft-same-256-128] (0005_branch-)             355.6709 (3.45)   
test_convolve1d_gpu[fft-same-256-128] (0004_updates)             357.0666 (3.47)   
test_convolve1d_gpu[fft-full-256-1025] (0004_updates)            360.4889 (3.50)   
test_convolve1d_gpu[fft-full-256-1025] (0005_branch-)            362.9680 (3.52)   
test_convolve1d_gpu[fft-valid-125-128] (0004_updates)            363.6082 (3.53)   
test_convolve1d_gpu[fft-valid-125-128] (0005_branch-)            363.7714 (3.53)   
test_convolve1d_gpu[fft-valid-256-128] (0004_updates)            364.2309 (3.54)   
test_convolve1d_gpu[fft-full-125-1025] (0004_updates)            365.5440 (3.55)   
test_convolve1d_gpu[fft-valid-256-128] (0005_branch-)            365.8781 (3.55)   
test_convolve1d_gpu[auto-same-256-128] (0004_updates)            370.9369 (3.60)   
test_convolve1d_gpu[fft-full-8192-1025] (0005_branch-)           371.0664 (3.60)   
test_convolve1d_gpu[fft-full-8192-1025] (0004_updates)           371.1718 (3.60)   
test_convolve1d_gpu[fft-full-8192-128] (0005_branch-)            371.5301 (3.61)   
test_convolve1d_gpu[auto-same-256-128] (0005_branch-)            372.9868 (3.62)   
test_convolve1d_gpu[fft-full-125-1025] (0005_branch-)            373.9818 (3.63)   
test_convolve1d_gpu[fft-full-125-8192] (0005_branch-)            374.1485 (3.63)   
test_convolve1d_gpu[fft-full-125-8192] (0004_updates)            374.4533 (3.64)   
test_convolve1d_gpu[fft-full-256-8192] (0005_branch-)            374.5713 (3.64)   
test_convolve1d_gpu[fft-full-256-8192] (0004_updates)            375.8082 (3.65)   
test_convolve1d_gpu[direct-same-8192-1025] (0004_updates)        376.6569 (3.66)   
test_convolve1d_gpu[fft-full-8192-8192] (0005_branch-)           376.7981 (3.66)   
test_convolve1d_gpu[fft-full-8192-128] (0004_updates)            377.3188 (3.66)   
test_convolve1d_gpu[fft-same-256-1025] (0004_updates)            378.1525 (3.67)   
test_convolve1d_gpu[fft-same-256-1025] (0005_branch-)            379.3248 (3.68)   
test_convolve1d_gpu[fft-same-125-1025] (0004_updates)            380.0194 (3.69)   
test_convolve1d_gpu[fft-full-8192-8192] (0004_updates)           380.8654 (3.70)   
test_convolve1d_gpu[fft-same-8192-1025] (0005_branch-)           381.4188 (3.70)   
test_convolve1d_gpu[fft-same-125-1025] (0005_branch-)            381.5330 (3.70)   
test_convolve1d_gpu[auto-full-256-1025] (0005_branch-)           382.8616 (3.72)   
test_convolve1d_gpu[direct-valid-8192-1025] (0005_branch-)       385.3879 (3.74)   
test_convolve1d_gpu[direct-valid-8192-1025] (0004_updates)       386.5493 (3.75)   
test_convolve1d_gpu[fft-same-125-8192] (0005_branch-)            387.2259 (3.76)   
test_convolve1d_gpu[fft-valid-125-1025] (0005_branch-)           387.5006 (3.76)   
test_convolve1d_gpu[fft-same-256-8192] (0005_branch-)            388.2841 (3.77)   
test_convolve1d_gpu[fft-valid-125-1025] (0004_updates)           388.8093 (3.77)   
test_convolve1d_gpu[fft-same-8192-128] (0005_branch-)            388.8176 (3.77)   
test_convolve1d_gpu[fft-valid-256-1025] (0005_branch-)           389.1044 (3.78)   
test_convolve1d_gpu[direct-valid-8192-8192] (0005_branch-)       389.7186 (3.78)   
test_convolve1d_gpu[auto-full-256-1025] (0004_updates)           390.3051 (3.79)   
test_convolve1d_gpu[fft-valid-256-1025] (0004_updates)           390.3539 (3.79)   
test_convolve1d_gpu[auto-same-125-1025] (0004_updates)           390.8209 (3.79)   
test_convolve1d_gpu[direct-valid-8192-8192] (0004_updates)       390.9154 (3.80)   
test_convolve1d_gpu[fft-same-8192-1025] (0004_updates)           391.7301 (3.80)   
test_convolve1d_gpu[auto-full-125-8192] (0005_branch-)           392.3763 (3.81)   
test_convolve1d_gpu[fft-same-256-8192] (0004_updates)            393.1001 (3.82)   
test_convolve1d_gpu[fft-same-125-8192] (0004_updates)            393.1109 (3.82)   
test_convolve1d_gpu[auto-same-256-1025] (0004_updates)           394.1816 (3.83)   
test_convolve1d_gpu[auto-full-125-8192] (0004_updates)           395.3009 (3.84)   
test_convolve1d_gpu[auto-full-8192-1025] (0004_updates)          397.3300 (3.86)   
test_convolve1d_gpu[fft-same-8192-8192] (0005_branch-)           397.5101 (3.86)   
test_convolve1d_gpu[direct-same-8192-1025] (0005_branch-)        397.8138 (3.86)   
test_convolve1d_gpu[auto-same-256-1025] (0005_branch-)           398.0882 (3.86)   
test_convolve1d_gpu[fft-same-8192-128] (0004_updates)            398.3255 (3.87)   
test_convolve1d_gpu[fft-valid-8192-1025] (0004_updates)          398.3275 (3.87)   
test_convolve1d_gpu[direct-full-8192-1025] (0004_updates)        398.5729 (3.87)   
test_convolve1d_gpu[auto-full-8192-128] (0005_branch-)           398.7149 (3.87)   
test_convolve1d_gpu[direct-full-8192-1025] (0005_branch-)        399.3824 (3.88)   
test_convolve1d_gpu[auto-same-125-1025] (0005_branch-)           399.7034 (3.88)   
test_convolve1d_gpu[fft-valid-125-8192] (0005_branch-)           400.4972 (3.89)   
test_convolve1d_gpu[fft-valid-8192-1025] (0005_branch-)          400.7795 (3.89)   
test_convolve1d_gpu[fft-valid-256-8192] (0005_branch-)           402.1392 (3.90)   
test_convolve1d_gpu[auto-same-8192-1025] (0005_branch-)          402.9592 (3.91)   
test_convolve1d_gpu[auto-full-256-8192] (0004_updates)           403.2025 (3.91)   
test_convolve1d_gpu[auto-same-125-8192] (0004_updates)           403.3266 (3.92)   
test_convolve1d_gpu[fft-valid-8192-128] (0004_updates)           404.6614 (3.93)   
test_convolve1d_gpu[auto-full-8192-128] (0004_updates)           404.8052 (3.93)   
test_convolve1d_gpu[auto-same-8192-128] (0005_branch-)           404.8331 (3.93)   
test_convolve1d_gpu[auto-same-8192-128] (0004_updates)           404.8593 (3.93)   
test_convolve1d_gpu[fft-valid-256-8192] (0004_updates)           405.0673 (3.93)   
test_convolve1d_gpu[auto-valid-8192-8192] (0004_updates)         406.2525 (3.94)   
test_convolve1d_gpu[fft-same-8192-8192] (0004_updates)           406.3658 (3.95)   
test_convolve1d_gpu[auto-full-8192-8192] (0004_updates)          407.5750 (3.96)   
test_convolve1d_gpu[fft-valid-8192-128] (0005_branch-)           407.9454 (3.96)   
test_convolve1d_gpu[auto-same-256-8192] (0005_branch-)           408.3274 (3.96)   
test_convolve1d_gpu[fft-valid-8192-8192] (0005_branch-)          409.4038 (3.97)   
test_convolve1d_gpu[fft-valid-125-8192] (0004_updates)           409.4135 (3.97)   
test_convolve1d_gpu[auto-same-125-8192] (0005_branch-)           410.8741 (3.99)   
test_convolve1d_gpu[auto-valid-8192-8192] (0005_branch-)         411.7055 (4.00)   
test_convolve1d_gpu[auto-same-8192-1025] (0004_updates)          411.8615 (4.00)   
test_convolve1d_gpu[auto-valid-8192-1025] (0004_updates)         413.0875 (4.01)   
test_convolve1d_gpu[auto-same-8192-8192] (0004_updates)          413.3094 (4.01)   
test_convolve1d_gpu[auto-same-256-8192] (0004_updates)           413.4741 (4.01)   
test_convolve1d_gpu[auto-valid-8192-1025] (0005_branch-)         417.1811 (4.05)   
test_convolve1d_gpu[auto-valid-256-8192] (0004_updates)          418.7262 (4.07)   
test_convolve1d_gpu[auto-same-8192-8192] (0005_branch-)          419.1619 (4.07)   
test_convolve1d_gpu[fft-valid-8192-8192] (0004_updates)          421.8060 (4.10)   
test_convolve1d_gpu[auto-valid-8192-128] (0004_updates)          421.8571 (4.10)   
test_convolve1d_gpu[auto-full-256-8192] (0005_branch-)           423.5274 (4.11)   
test_convolve1d_gpu[auto-valid-8192-128] (0005_branch-)          425.0145 (4.13)   
test_convolve1d_gpu[auto-valid-125-8192] (0004_updates)          429.4510 (4.17)   
test_convolve1d_gpu[auto-valid-256-8192] (0005_branch-)          430.0432 (4.18)   
test_convolve1d_gpu[auto-valid-125-8192] (0005_branch-)          444.4414 (4.31)   
test_convolve1d_gpu[auto-full-8192-8192] (0005_branch-)          606.1195 (5.88)   
test_convolve1d_gpu[auto-full-8192-1025] (0005_branch-)          815.4478 (7.92)   
test_convolve1d_gpu[direct-same-8192-8192] (0004_updates)      2,488.0675 (24.16)  
test_convolve1d_gpu[direct-same-8192-8192] (0005_branch-)      2,519.4995 (24.46)  
test_convolve1d_gpu[direct-full-8192-8192] (0004_updates)      2,561.9181 (24.87)  
test_convolve1d_gpu[direct-full-8192-8192] (0005_branch-)      2,575.8184 (25.01)  
-----------------------------------------------------------------------------------


----------------------- benchmark 'Convolve2d': 36 tests ----------------------
Name (time in us)                                                Mean          
-------------------------------------------------------------------------------
test_convolve2d_gpu[valid-wrap-5-256] (0004_updates)         112.1240 (1.0)    
test_convolve2d_gpu[valid-symm-5-256] (0004_updates)         112.4648 (1.00)   
test_convolve2d_gpu[valid-fill-5-256] (0004_updates)         116.6409 (1.04)   
test_convolve2d_gpu[valid-wrap-5-256] (0005_branch-)         117.2213 (1.05)   
test_convolve2d_gpu[valid-symm-5-256] (0005_branch-)         117.7130 (1.05)   
test_convolve2d_gpu[valid-fill-5-256] (0005_branch-)         121.3992 (1.08)   
test_convolve2d_gpu[same-fill-5-256] (0004_updates)          224.5570 (2.00)   
test_convolve2d_gpu[full-fill-5-256] (0004_updates)          229.6875 (2.05)   
test_convolve2d_gpu[same-fill-5-256] (0005_branch-)          234.8044 (2.09)   
test_convolve2d_gpu[full-fill-5-256] (0005_branch-)          238.5044 (2.13)   
test_convolve2d_gpu[same-wrap-5-256] (0004_updates)          254.7049 (2.27)   
test_convolve2d_gpu[same-symm-5-256] (0004_updates)          257.0184 (2.29)   
test_convolve2d_gpu[full-wrap-5-256] (0004_updates)          259.7478 (2.32)   
test_convolve2d_gpu[full-symm-5-256] (0004_updates)          264.4727 (2.36)   
test_convolve2d_gpu[same-wrap-5-256] (0005_branch-)          266.7005 (2.38)   
test_convolve2d_gpu[full-wrap-5-256] (0005_branch-)          267.4655 (2.39)   
test_convolve2d_gpu[same-symm-5-256] (0005_branch-)          271.8703 (2.42)   
test_convolve2d_gpu[full-symm-5-256] (0005_branch-)          279.3825 (2.49)   
test_convolve2d_gpu[valid-wrap-100-256] (0004_updates)     1,604.1086 (14.31)  
test_convolve2d_gpu[valid-symm-100-256] (0004_updates)     1,605.6558 (14.32)  
test_convolve2d_gpu[valid-wrap-100-256] (0005_branch-)     1,614.0480 (14.40)  
test_convolve2d_gpu[valid-symm-100-256] (0005_branch-)     1,614.7143 (14.40)  
test_convolve2d_gpu[valid-fill-100-256] (0004_updates)     1,683.7307 (15.02)  
test_convolve2d_gpu[valid-fill-100-256] (0005_branch-)     1,735.3781 (15.48)  
test_convolve2d_gpu[same-fill-100-256] (0004_updates)      3,199.0321 (28.53)  
test_convolve2d_gpu[same-fill-100-256] (0005_branch-)      3,225.7229 (28.77)  
test_convolve2d_gpu[same-symm-100-256] (0004_updates)      3,270.7409 (29.17)  
test_convolve2d_gpu[same-symm-100-256] (0005_branch-)      3,351.0881 (29.89)  
test_convolve2d_gpu[same-wrap-100-256] (0004_updates)      3,380.0068 (30.15)  
test_convolve2d_gpu[same-wrap-100-256] (0005_branch-)      3,441.5133 (30.69)  
test_convolve2d_gpu[full-fill-100-256] (0004_updates)      6,118.7811 (54.57)  
test_convolve2d_gpu[full-symm-100-256] (0005_branch-)      6,189.1810 (55.20)  
test_convolve2d_gpu[full-symm-100-256] (0004_updates)      6,267.7008 (55.90)  
test_convolve2d_gpu[full-wrap-100-256] (0005_branch-)      6,278.8711 (56.00)  
test_convolve2d_gpu[full-wrap-100-256] (0004_updates)      6,328.6094 (56.44)  
test_convolve2d_gpu[full-fill-100-256] (0005_branch-)      6,546.6531 (58.39)  
-------------------------------------------------------------------------------


------------------------- benchmark 'Correlate': 162 tests -------------------------
Name (time in us)                                                     Mean          
------------------------------------------------------------------------------------
test_correlate1d_gpu[direct-same-8192-128] (0004_updates)         101.9377 (1.0)    
test_correlate1d_gpu[direct-same-125-128] (0005_branch-)          102.9257 (1.01)   
test_correlate1d_gpu[direct-same-256-128] (0004_updates)          103.1625 (1.01)   
test_correlate1d_gpu[direct-same-256-128] (0005_branch-)          103.4191 (1.01)   
test_correlate1d_gpu[direct-same-125-128] (0004_updates)          103.8374 (1.02)   
test_correlate1d_gpu[direct-same-8192-128] (0005_branch-)         103.9006 (1.02)   
test_correlate1d_gpu[direct-valid-125-128] (0004_updates)         104.3522 (1.02)   
test_correlate1d_gpu[direct-full-125-128] (0004_updates)          105.6878 (1.04)   
test_correlate1d_gpu[direct-valid-125-128] (0005_branch-)         105.8682 (1.04)   
test_correlate1d_gpu[direct-valid-256-128] (0004_updates)         106.6431 (1.05)   
test_correlate1d_gpu[direct-valid-256-128] (0005_branch-)         109.0206 (1.07)   
test_correlate1d_gpu[direct-full-125-128] (0005_branch-)          110.8210 (1.09)   
test_correlate1d_gpu[direct-full-256-128] (0004_updates)          112.6627 (1.11)   
test_correlate1d_gpu[direct-full-256-128] (0005_branch-)          116.4907 (1.14)   
test_correlate1d_gpu[auto-same-125-128] (0005_branch-)            128.0247 (1.26)   
test_correlate1d_gpu[auto-same-125-128] (0004_updates)            130.1547 (1.28)   
test_correlate1d_gpu[direct-same-125-1025] (0005_branch-)         131.0149 (1.29)   
test_correlate1d_gpu[auto-full-125-128] (0004_updates)            132.1208 (1.30)   
test_correlate1d_gpu[direct-same-125-1025] (0004_updates)         132.6685 (1.30)   
test_correlate1d_gpu[direct-valid-125-1025] (0004_updates)        132.6735 (1.30)   
test_correlate1d_gpu[direct-full-8192-128] (0005_branch-)         133.1852 (1.31)   
test_correlate1d_gpu[auto-valid-125-128] (0004_updates)           133.2475 (1.31)   
test_correlate1d_gpu[auto-valid-256-128] (0005_branch-)           133.7815 (1.31)   
test_correlate1d_gpu[auto-valid-125-128] (0005_branch-)           133.7961 (1.31)   
test_correlate1d_gpu[direct-valid-8192-128] (0005_branch-)        133.8214 (1.31)   
test_correlate1d_gpu[direct-valid-125-1025] (0005_branch-)        134.0143 (1.31)   
test_correlate1d_gpu[auto-valid-256-128] (0004_updates)           134.6865 (1.32)   
test_correlate1d_gpu[auto-full-125-128] (0005_branch-)            134.9694 (1.32)   
test_correlate1d_gpu[direct-valid-8192-128] (0004_updates)        135.0451 (1.32)   
test_correlate1d_gpu[direct-same-125-8192] (0004_updates)         138.5731 (1.36)   
test_correlate1d_gpu[direct-same-125-8192] (0005_branch-)         138.7724 (1.36)   
test_correlate1d_gpu[direct-full-125-1025] (0004_updates)         139.9243 (1.37)   
test_correlate1d_gpu[auto-full-256-128] (0004_updates)            141.4478 (1.39)   
test_correlate1d_gpu[direct-full-125-8192] (0004_updates)         142.2016 (1.39)   
test_correlate1d_gpu[direct-valid-125-8192] (0004_updates)        142.8819 (1.40)   
test_correlate1d_gpu[auto-full-256-128] (0005_branch-)            143.4930 (1.41)   
test_correlate1d_gpu[direct-full-8192-128] (0004_updates)         144.7058 (1.42)   
test_correlate1d_gpu[direct-full-125-8192] (0005_branch-)         144.9224 (1.42)   
test_correlate1d_gpu[direct-valid-125-8192] (0005_branch-)        145.0260 (1.42)   
test_correlate1d_gpu[direct-full-125-1025] (0005_branch-)         146.1195 (1.43)   
test_correlate1d_gpu[direct-valid-256-1025] (0004_updates)        160.9474 (1.58)   
test_correlate1d_gpu[direct-valid-256-1025] (0005_branch-)        164.3962 (1.61)   
test_correlate1d_gpu[direct-same-256-8192] (0004_updates)         165.6696 (1.63)   
test_correlate1d_gpu[direct-full-256-8192] (0004_updates)         165.6830 (1.63)   
test_correlate1d_gpu[direct-same-256-8192] (0005_branch-)         167.8565 (1.65)   
test_correlate1d_gpu[direct-valid-256-8192] (0005_branch-)        168.9237 (1.66)   
test_correlate1d_gpu[direct-valid-256-8192] (0004_updates)        169.3692 (1.66)   
test_correlate1d_gpu[direct-full-256-8192] (0005_branch-)         169.6886 (1.66)   
test_correlate1d_gpu[direct-full-256-1025] (0004_updates)         172.3371 (1.69)   
test_correlate1d_gpu[direct-same-256-1025] (0004_updates)         172.8284 (1.70)   
test_correlate1d_gpu[auto-valid-125-1025] (0004_updates)          173.7643 (1.70)   
test_correlate1d_gpu[auto-full-125-1025] (0004_updates)           173.8500 (1.71)   
test_correlate1d_gpu[direct-full-256-1025] (0005_branch-)         174.9858 (1.72)   
test_correlate1d_gpu[auto-valid-125-1025] (0005_branch-)          175.4383 (1.72)   
test_correlate1d_gpu[direct-same-256-1025] (0005_branch-)         176.6022 (1.73)   
test_correlate1d_gpu[auto-full-125-1025] (0005_branch-)           176.9355 (1.74)   
test_correlate1d_gpu[auto-valid-256-1025] (0005_branch-)          201.9720 (1.98)   
test_correlate1d_gpu[auto-valid-256-1025] (0004_updates)          202.8147 (1.99)   
test_correlate1d_gpu[fft-full-125-128] (0005_branch-)             357.8518 (3.51)   
test_correlate1d_gpu[fft-full-256-128] (0005_branch-)             359.8996 (3.53)   
test_correlate1d_gpu[fft-full-125-128] (0004_updates)             365.6261 (3.59)   
test_correlate1d_gpu[fft-full-256-128] (0004_updates)             368.7343 (3.62)   
test_correlate1d_gpu[fft-same-125-128] (0004_updates)             374.3814 (3.67)   
test_correlate1d_gpu[fft-same-125-128] (0005_branch-)             375.4125 (3.68)   
test_correlate1d_gpu[fft-same-256-128] (0004_updates)             376.7053 (3.70)   
test_correlate1d_gpu[direct-same-8192-1025] (0005_branch-)        377.3964 (3.70)   
test_correlate1d_gpu[direct-same-8192-1025] (0004_updates)        378.9122 (3.72)   
test_correlate1d_gpu[direct-valid-8192-1025] (0004_updates)       382.4669 (3.75)   
test_correlate1d_gpu[direct-valid-8192-1025] (0005_branch-)       383.4648 (3.76)   
test_correlate1d_gpu[fft-same-256-128] (0005_branch-)             384.4736 (3.77)   
test_correlate1d_gpu[fft-valid-125-128] (0004_updates)            385.4488 (3.78)   
test_correlate1d_gpu[fft-full-256-1025] (0004_updates)            387.3874 (3.80)   
test_correlate1d_gpu[fft-full-256-1025] (0005_branch-)            387.8229 (3.80)   
test_correlate1d_gpu[direct-valid-8192-8192] (0004_updates)       388.5044 (3.81)   
test_correlate1d_gpu[direct-valid-8192-8192] (0005_branch-)       389.2859 (3.82)   
test_correlate1d_gpu[fft-full-125-1025] (0004_updates)            390.2637 (3.83)   
test_correlate1d_gpu[fft-full-125-1025] (0005_branch-)            390.9833 (3.84)   
test_correlate1d_gpu[fft-valid-256-128] (0004_updates)            391.2555 (3.84)   
test_correlate1d_gpu[fft-valid-125-128] (0005_branch-)            391.4486 (3.84)   
test_correlate1d_gpu[auto-same-256-128] (0005_branch-)            392.5752 (3.85)   
test_correlate1d_gpu[fft-full-8192-1025] (0004_updates)           392.6367 (3.85)   
test_correlate1d_gpu[fft-valid-256-128] (0005_branch-)            393.1657 (3.86)   
test_correlate1d_gpu[fft-full-256-8192] (0004_updates)            396.2097 (3.89)   
test_correlate1d_gpu[fft-full-8192-1025] (0005_branch-)           396.5232 (3.89)   
test_correlate1d_gpu[fft-full-8192-128] (0004_updates)            397.7181 (3.90)   
test_correlate1d_gpu[direct-full-8192-1025] (0004_updates)        397.7971 (3.90)   
test_correlate1d_gpu[fft-same-256-1025] (0004_updates)            397.9318 (3.90)   
test_correlate1d_gpu[fft-full-256-8192] (0005_branch-)            398.1188 (3.91)   
test_correlate1d_gpu[direct-full-8192-1025] (0005_branch-)        399.2387 (3.92)   
test_correlate1d_gpu[fft-same-125-1025] (0004_updates)            399.6199 (3.92)   
test_correlate1d_gpu[fft-full-8192-8192] (0005_branch-)           401.4529 (3.94)   
test_correlate1d_gpu[fft-full-8192-128] (0005_branch-)            401.4552 (3.94)   
test_correlate1d_gpu[fft-full-125-8192] (0005_branch-)            401.8826 (3.94)   
test_correlate1d_gpu[auto-same-256-128] (0004_updates)            402.5471 (3.95)   
test_correlate1d_gpu[fft-same-125-1025] (0005_branch-)            403.8391 (3.96)   
test_correlate1d_gpu[fft-same-256-1025] (0005_branch-)            404.5995 (3.97)   
test_correlate1d_gpu[auto-full-256-1025] (0004_updates)           406.3415 (3.99)   
test_correlate1d_gpu[auto-full-256-1025] (0005_branch-)           407.4092 (4.00)   
test_correlate1d_gpu[fft-same-8192-1025] (0004_updates)           409.8067 (4.02)   
test_correlate1d_gpu[fft-same-8192-128] (0004_updates)            411.2330 (4.03)   
test_correlate1d_gpu[fft-valid-256-1025] (0004_updates)           412.5848 (4.05)   
test_correlate1d_gpu[fft-same-8192-1025] (0005_branch-)           412.7209 (4.05)   
test_correlate1d_gpu[fft-valid-125-1025] (0004_updates)           414.3900 (4.07)   
test_correlate1d_gpu[fft-valid-256-1025] (0005_branch-)           414.7512 (4.07)   
test_correlate1d_gpu[fft-valid-125-1025] (0005_branch-)           415.1777 (4.07)   
test_correlate1d_gpu[fft-same-256-8192] (0004_updates)            415.6919 (4.08)   
test_correlate1d_gpu[fft-same-256-8192] (0005_branch-)            415.7055 (4.08)   
test_correlate1d_gpu[auto-valid-8192-8192] (0004_updates)         415.9500 (4.08)   
test_correlate1d_gpu[fft-same-125-8192] (0005_branch-)            416.4651 (4.09)   
test_correlate1d_gpu[fft-same-125-8192] (0004_updates)            416.5472 (4.09)   
test_correlate1d_gpu[fft-full-125-8192] (0004_updates)            417.0435 (4.09)   
test_correlate1d_gpu[auto-full-8192-1025] (0004_updates)          417.5374 (4.10)   
test_correlate1d_gpu[fft-same-8192-128] (0005_branch-)            417.9394 (4.10)   
test_correlate1d_gpu[auto-valid-8192-8192] (0005_branch-)         418.0794 (4.10)   
test_correlate1d_gpu[auto-same-256-1025] (0005_branch-)           419.3367 (4.11)   
test_correlate1d_gpu[auto-same-125-1025] (0004_updates)           419.8635 (4.12)   
test_correlate1d_gpu[auto-full-125-8192] (0004_updates)           420.4320 (4.12)   
test_correlate1d_gpu[auto-full-8192-128] (0005_branch-)           420.6969 (4.13)   
test_correlate1d_gpu[fft-full-8192-8192] (0004_updates)           421.7962 (4.14)   
test_correlate1d_gpu[auto-full-256-8192] (0005_branch-)           421.9281 (4.14)   
test_correlate1d_gpu[auto-full-8192-128] (0004_updates)           422.5889 (4.15)   
test_correlate1d_gpu[fft-valid-8192-1025] (0005_branch-)          422.9191 (4.15)   
test_correlate1d_gpu[fft-valid-8192-128] (0004_updates)           423.5746 (4.16)   
test_correlate1d_gpu[auto-same-125-1025] (0005_branch-)           423.6915 (4.16)   
test_correlate1d_gpu[fft-valid-125-8192] (0004_updates)           423.9364 (4.16)   
test_correlate1d_gpu[fft-valid-8192-1025] (0004_updates)          424.3427 (4.16)   
test_correlate1d_gpu[auto-full-256-8192] (0004_updates)           424.4003 (4.16)   
test_correlate1d_gpu[fft-same-8192-8192] (0004_updates)           424.7035 (4.17)   
test_correlate1d_gpu[fft-same-8192-8192] (0005_branch-)           426.2820 (4.18)   
test_correlate1d_gpu[fft-valid-125-8192] (0005_branch-)           426.6273 (4.19)   
test_correlate1d_gpu[auto-full-8192-8192] (0005_branch-)          426.9976 (4.19)   
test_correlate1d_gpu[fft-valid-256-8192] (0004_updates)           427.3843 (4.19)   
test_correlate1d_gpu[auto-same-256-8192] (0005_branch-)           428.9800 (4.21)   
test_correlate1d_gpu[fft-valid-8192-8192] (0005_branch-)          429.3137 (4.21)   
test_correlate1d_gpu[auto-same-256-1025] (0004_updates)           429.6409 (4.21)   
test_correlate1d_gpu[auto-same-8192-1025] (0004_updates)          430.8967 (4.23)   
test_correlate1d_gpu[auto-same-8192-128] (0005_branch-)           430.9231 (4.23)   
test_correlate1d_gpu[fft-valid-256-8192] (0005_branch-)           431.2052 (4.23)   
test_correlate1d_gpu[auto-same-125-8192] (0005_branch-)           431.4352 (4.23)   
test_correlate1d_gpu[auto-full-125-8192] (0005_branch-)           432.0364 (4.24)   
test_correlate1d_gpu[fft-valid-8192-128] (0005_branch-)           432.1660 (4.24)   
test_correlate1d_gpu[auto-same-8192-1025] (0005_branch-)          432.5029 (4.24)   
test_correlate1d_gpu[auto-full-8192-8192] (0004_updates)          433.5697 (4.25)   
test_correlate1d_gpu[fft-valid-8192-8192] (0004_updates)          433.9593 (4.26)   
test_correlate1d_gpu[auto-same-256-8192] (0004_updates)           435.5418 (4.27)   
test_correlate1d_gpu[auto-same-8192-128] (0004_updates)           435.8881 (4.28)   
test_correlate1d_gpu[auto-same-125-8192] (0004_updates)           436.6123 (4.28)   
test_correlate1d_gpu[auto-valid-8192-1025] (0005_branch-)         440.1744 (4.32)   
test_correlate1d_gpu[auto-same-8192-8192] (0005_branch-)          444.4949 (4.36)   
test_correlate1d_gpu[auto-same-8192-8192] (0004_updates)          445.2173 (4.37)   
test_correlate1d_gpu[auto-valid-256-8192] (0005_branch-)          447.1757 (4.39)   
test_correlate1d_gpu[auto-valid-8192-128] (0005_branch-)          448.3286 (4.40)   
test_correlate1d_gpu[auto-valid-125-8192] (0005_branch-)          450.9786 (4.42)   
test_correlate1d_gpu[auto-valid-8192-128] (0004_updates)          452.5396 (4.44)   
test_correlate1d_gpu[auto-full-8192-1025] (0005_branch-)          453.1844 (4.45)   
test_correlate1d_gpu[auto-valid-8192-1025] (0004_updates)         454.0802 (4.45)   
test_correlate1d_gpu[auto-valid-125-8192] (0004_updates)          454.5219 (4.46)   
test_correlate1d_gpu[auto-valid-256-8192] (0004_updates)          458.3873 (4.50)   
test_correlate1d_gpu[direct-same-8192-8192] (0004_updates)      2,473.9599 (24.27)  
test_correlate1d_gpu[direct-same-8192-8192] (0005_branch-)      2,482.0718 (24.35)  
test_correlate1d_gpu[direct-full-8192-8192] (0005_branch-)      2,557.5257 (25.09)  
test_correlate1d_gpu[direct-full-8192-8192] (0004_updates)      2,566.4860 (25.18)  
------------------------------------------------------------------------------------


---------------------- benchmark 'Correlate2d': 36 tests -----------------------
Name (time in us)                                                 Mean          
--------------------------------------------------------------------------------
test_correlate2d_gpu[valid-symm-5-256] (0004_updates)         111.7596 (1.0)    
test_correlate2d_gpu[valid-wrap-5-256] (0004_updates)         115.3436 (1.03)   
test_correlate2d_gpu[valid-fill-5-256] (0004_updates)         117.3263 (1.05)   
test_correlate2d_gpu[valid-wrap-5-256] (0005_branch-)         118.0128 (1.06)   
test_correlate2d_gpu[valid-symm-5-256] (0005_branch-)         119.0308 (1.07)   
test_correlate2d_gpu[valid-fill-5-256] (0005_branch-)         122.2320 (1.09)   
test_correlate2d_gpu[same-fill-5-256] (0004_updates)          223.7750 (2.00)   
test_correlate2d_gpu[full-fill-5-256] (0004_updates)          226.6957 (2.03)   
test_correlate2d_gpu[same-fill-5-256] (0005_branch-)          236.7009 (2.12)   
test_correlate2d_gpu[full-fill-5-256] (0005_branch-)          238.6407 (2.14)   
test_correlate2d_gpu[same-wrap-5-256] (0004_updates)          259.5734 (2.32)   
test_correlate2d_gpu[same-symm-5-256] (0004_updates)          261.0049 (2.34)   
test_correlate2d_gpu[full-symm-5-256] (0004_updates)          263.1936 (2.35)   
test_correlate2d_gpu[full-wrap-5-256] (0004_updates)          263.2617 (2.36)   
test_correlate2d_gpu[full-wrap-5-256] (0005_branch-)          267.0346 (2.39)   
test_correlate2d_gpu[same-symm-5-256] (0005_branch-)          267.6500 (2.39)   
test_correlate2d_gpu[same-wrap-5-256] (0005_branch-)          269.3223 (2.41)   
test_correlate2d_gpu[full-symm-5-256] (0005_branch-)          272.9880 (2.44)   
test_correlate2d_gpu[valid-wrap-100-256] (0004_updates)     1,600.7832 (14.32)  
test_correlate2d_gpu[valid-symm-100-256] (0004_updates)     1,602.4842 (14.34)  
test_correlate2d_gpu[valid-wrap-100-256] (0005_branch-)     1,607.6571 (14.38)  
test_correlate2d_gpu[valid-symm-100-256] (0005_branch-)     1,610.9417 (14.41)  
test_correlate2d_gpu[valid-fill-100-256] (0004_updates)     1,648.7860 (14.75)  
test_correlate2d_gpu[valid-fill-100-256] (0005_branch-)     1,706.1204 (15.27)  
test_correlate2d_gpu[same-fill-100-256] (0004_updates)      3,225.8889 (28.86)  
test_correlate2d_gpu[same-fill-100-256] (0005_branch-)      3,244.3383 (29.03)  
test_correlate2d_gpu[same-symm-100-256] (0004_updates)      3,293.5472 (29.47)  
test_correlate2d_gpu[same-symm-100-256] (0005_branch-)      3,353.3057 (30.00)  
test_correlate2d_gpu[same-wrap-100-256] (0004_updates)      3,426.1373 (30.66)  
test_correlate2d_gpu[same-wrap-100-256] (0005_branch-)      3,465.9630 (31.01)  
test_correlate2d_gpu[full-symm-100-256] (0004_updates)      6,141.1713 (54.95)  
test_correlate2d_gpu[full-symm-100-256] (0005_branch-)      6,202.5638 (55.50)  
test_correlate2d_gpu[full-wrap-100-256] (0005_branch-)      6,212.2288 (55.59)  
test_correlate2d_gpu[full-wrap-100-256] (0004_updates)      6,222.7590 (55.68)  
test_correlate2d_gpu[full-fill-100-256] (0005_branch-)      6,231.1092 (55.75)  
test_correlate2d_gpu[full-fill-100-256] (0004_updates)      6,309.2902 (56.45)  
--------------------------------------------------------------------------------


--------------- benchmark 'Cosine': 2 tests ---------------
Name (time in us)                            Mean          
-----------------------------------------------------------
test_cosine_gpu[32768] (0004_updates)     14.6284 (1.0)    
test_cosine_gpu[32768] (0005_branch-)     16.2895 (1.11)   
-----------------------------------------------------------


--------------- benchmark 'Cubic': 2 tests ---------------
Name (time in us)                           Mean          
----------------------------------------------------------
test_cubic_gpu[65536] (0004_updates)     13.9709 (1.0)    
test_cubic_gpu[65536] (0005_branch-)     14.8084 (1.06)   
----------------------------------------------------------


------------------------ benchmark 'Decimate': 60 tests ------------------------
Name (time in us)                                                 Mean          
--------------------------------------------------------------------------------
test_decimate_gpu[False-False-2-16384] (0004_updates)         357.7813 (1.0)    
test_decimate_gpu[False-True-2-16384] (0004_updates)          370.3604 (1.04)   
test_decimate_gpu[False-False-2-262144] (0004_updates)        372.5145 (1.04)   
test_decimate_gpu[False-False-3-16384] (0004_updates)         376.4401 (1.05)   
test_decimate_gpu[False-False-4-16384] (0004_updates)         378.7639 (1.06)   
test_decimate_gpu[False-True-2-262144] (0004_updates)         380.8322 (1.06)   
test_decimate_gpu[False-True-3-16384] (0004_updates)          387.9568 (1.08)   
test_decimate_gpu[False-False-3-262144] (0004_updates)        388.3685 (1.09)   
test_decimate_gpu[False-True-3-262144] (0004_updates)         395.4973 (1.11)   
test_decimate_gpu[False-True-4-16384] (0004_updates)          395.9325 (1.11)   
test_decimate_gpu[False-False-4-262144] (0004_updates)        397.8586 (1.11)   
test_decimate_gpu[False-True-4-262144] (0004_updates)         407.1185 (1.14)   
test_decimate_gpu[False-False-8-16384] (0004_updates)         429.5962 (1.20)   
test_decimate_gpu[False-False-8-262144] (0004_updates)        436.6332 (1.22)   
test_decimate_gpu[False-True-8-262144] (0004_updates)         445.8360 (1.25)   
test_decimate_gpu[False-True-8-16384] (0004_updates)          446.5001 (1.25)   
test_decimate_gpu[True-False-2-16384] (0004_updates)          544.4871 (1.52)   
test_decimate_gpu[True-False-2-262144] (0004_updates)         557.4987 (1.56)   
test_decimate_gpu[True-False-3-16384] (0004_updates)          561.6851 (1.57)   
test_decimate_gpu[True-False-3-262144] (0004_updates)         569.4387 (1.59)   
test_decimate_gpu[True-False-4-16384] (0004_updates)          573.9015 (1.60)   
test_decimate_gpu[True-False-4-262144] (0004_updates)         579.5083 (1.62)   
test_decimate_gpu[True-False-8-262144] (0004_updates)         611.3724 (1.71)   
test_decimate_gpu[True-False-8-16384] (0004_updates)          618.1679 (1.73)   
test_decimate_gpu[True-True-3-16384] (0004_updates)           626.2995 (1.75)   
test_decimate_gpu[True-True-4-16384] (0004_updates)           633.2300 (1.77)   
test_decimate_gpu[True-True-2-262144] (0004_updates)          634.4448 (1.77)   
test_decimate_gpu[True-True-3-262144] (0004_updates)          636.7869 (1.78)   
test_decimate_gpu[True-True-2-16384] (0004_updates)           637.2235 (1.78)   
test_decimate_gpu[True-True-4-262144] (0004_updates)          644.9696 (1.80)   
test_decimate_gpu[True-True-8-262144] (0004_updates)          676.1378 (1.89)   
test_decimate_gpu[False-False-64-16384] (0004_updates)        676.9441 (1.89)   
test_decimate_gpu[True-True-8-16384] (0004_updates)           680.1036 (1.90)   
test_decimate_gpu[False-True-64-16384] (0004_updates)         689.9335 (1.93)   
test_decimate_gpu[False-2-16384] (0005_branch-)               761.0704 (2.13)   
test_decimate_gpu[False-3-16384] (0005_branch-)               775.2829 (2.17)   
test_decimate_gpu[False-4-16384] (0005_branch-)               783.3174 (2.19)   
test_decimate_gpu[False-2-262144] (0005_branch-)              788.1444 (2.20)   
test_decimate_gpu[False-3-262144] (0005_branch-)              792.1745 (2.21)   
test_decimate_gpu[False-4-262144] (0005_branch-)              798.7386 (2.23)   
test_decimate_gpu[True-False-64-16384] (0004_updates)         816.7441 (2.28)   
test_decimate_gpu[False-8-16384] (0005_branch-)               828.7186 (2.32)   
test_decimate_gpu[True-2-16384] (0005_branch-)                833.6889 (2.33)   
test_decimate_gpu[True-3-16384] (0005_branch-)                838.6691 (2.34)   
test_decimate_gpu[True-4-16384] (0005_branch-)                845.1985 (2.36)   
test_decimate_gpu[False-8-262144] (0005_branch-)              845.8490 (2.36)   
test_decimate_gpu[True-2-262144] (0005_branch-)               855.9400 (2.39)   
test_decimate_gpu[True-3-262144] (0005_branch-)               862.4624 (2.41)   
test_decimate_gpu[True-4-262144] (0005_branch-)               863.9535 (2.41)   
test_decimate_gpu[True-8-16384] (0005_branch-)                897.5151 (2.51)   
test_decimate_gpu[True-8-262144] (0005_branch-)               900.3144 (2.52)   
test_decimate_gpu[True-True-64-16384] (0004_updates)          904.9318 (2.53)   
test_decimate_gpu[False-64-16384] (0005_branch-)            1,051.7363 (2.94)   
test_decimate_gpu[True-64-16384] (0005_branch-)             1,132.7332 (3.17)   
test_decimate_gpu[False-False-64-262144] (0004_updates)     1,253.3947 (3.50)   
test_decimate_gpu[False-True-64-262144] (0004_updates)      1,294.1023 (3.62)   
test_decimate_gpu[True-False-64-262144] (0004_updates)      1,391.6172 (3.89)   
test_decimate_gpu[True-True-64-262144] (0004_updates)       1,515.6263 (4.24)   
test_decimate_gpu[False-64-262144] (0005_branch-)           1,633.7275 (4.57)   
test_decimate_gpu[True-64-262144] (0005_branch-)            1,759.1275 (4.92)   
--------------------------------------------------------------------------------


---------------- benchmark 'Detrend': 2 tests ---------------
Name (time in us)                              Mean          
-------------------------------------------------------------
test_detrend_gpu[256] (0004_updates)       594.2241 (1.0)    
test_detrend_gpu[256] (0005_branch-)     1,348.2214 (2.27)   
-------------------------------------------------------------


----------------- benchmark 'Exponential': 4 tests -----------------
Name (time in us)                                     Mean          
--------------------------------------------------------------------
test_exponential_gpu[1.5-32768] (0004_updates)     16.5851 (1.0)    
test_exponential_gpu[3.0-32768] (0004_updates)     17.2066 (1.04)   
test_exponential_gpu[1.5-32768] (0005_branch-)     17.3536 (1.05)   
test_exponential_gpu[3.0-32768] (0005_branch-)     18.4711 (1.11)   
--------------------------------------------------------------------


------------------- benchmark 'FFTConvolve': 6 tests ------------------
Name (time in us)                                        Mean          
-----------------------------------------------------------------------
test_fftconvolve_gpu[full-32768] (0004_updates)      373.5981 (1.0)    
test_fftconvolve_gpu[same-32768] (0004_updates)      382.5187 (1.02)   
test_fftconvolve_gpu[full-32768] (0005_branch-)      384.3293 (1.03)   
test_fftconvolve_gpu[valid-32768] (0004_updates)     396.8504 (1.06)   
test_fftconvolve_gpu[same-32768] (0005_branch-)      398.8080 (1.07)   
test_fftconvolve_gpu[valid-32768] (0005_branch-)     417.1946 (1.12)   
-----------------------------------------------------------------------


-------------------- benchmark 'FirWin': 8 tests --------------------
Name (time in us)                                      Mean          
---------------------------------------------------------------------
test_firwin_gpu[0.2-0.1-32768] (0004_updates)      369.0950 (1.0)    
test_firwin_gpu[0.4-0.1-32768] (0004_updates)      372.2428 (1.01)   
test_firwin_gpu[0.2-0.15-32768] (0004_updates)     375.9900 (1.02)   
test_firwin_gpu[0.4-0.15-32768] (0004_updates)     377.2522 (1.02)   
test_firwin_gpu[0.4-0.1-32768] (0005_branch-)      630.0338 (1.71)   
test_firwin_gpu[0.2-0.1-32768] (0005_branch-)      630.0911 (1.71)   
test_firwin_gpu[0.4-0.15-32768] (0005_branch-)     631.3302 (1.71)   
test_firwin_gpu[0.2-0.15-32768] (0005_branch-)     637.4218 (1.73)   
---------------------------------------------------------------------


----------------- benchmark 'Firfilter': 12 tests ------------------
Name (time in us)                                     Mean          
--------------------------------------------------------------------
test_firfilter_gpu[128-16384] (0004_updates)      373.3141 (1.0)    
test_firfilter_gpu[32-16384] (0004_updates)       374.7543 (1.00)   
test_firfilter_gpu[8-16384] (0004_updates)        382.0688 (1.02)   
test_firfilter_gpu[128-16384] (0005_branch-)      386.4194 (1.04)   
test_firfilter_gpu[32-16384] (0005_branch-)       388.5247 (1.04)   
test_firfilter_gpu[8-16384] (0005_branch-)        396.5624 (1.06)   
test_firfilter_gpu[32-262144] (0004_updates)      529.4318 (1.42)   
test_firfilter_gpu[8-262144] (0004_updates)       529.6774 (1.42)   
test_firfilter_gpu[128-262144] (0004_updates)     530.1291 (1.42)   
test_firfilter_gpu[128-262144] (0005_branch-)     530.9050 (1.42)   
test_firfilter_gpu[8-262144] (0005_branch-)       531.3490 (1.42)   
test_firfilter_gpu[32-262144] (0005_branch-)      532.3393 (1.43)   
--------------------------------------------------------------------


--------------- benchmark 'FlatTop': 2 tests ---------------
Name (time in us)                             Mean          
------------------------------------------------------------
test_flattop_gpu[32768] (0004_updates)     64.1398 (1.0)    
test_flattop_gpu[32768] (0005_branch-)     66.3272 (1.03)   
------------------------------------------------------------


------------------------ benchmark 'FreqShift': 40 tests ------------------------
Name (time in us)                                                  Mean          
---------------------------------------------------------------------------------
test_freq_shift_gpu[0.3-0.0-256-float64] (0004_updates)         14.9764 (1.0)    
test_freq_shift_gpu[0.3-0.0-256-complex128] (0004_updates)      15.7276 (1.05)   
test_freq_shift_gpu[0.3--4.0-256-float64] (0004_updates)        16.5345 (1.10)   
test_freq_shift_gpu[0.3--5.0-256-float64] (0004_updates)        16.5584 (1.11)   
test_freq_shift_gpu[0.3-4.0-256-float64] (0004_updates)         16.5782 (1.11)   
test_freq_shift_gpu[0.3--2.0-256-float64] (0004_updates)        16.5864 (1.11)   
test_freq_shift_gpu[0.3--1.0-256-float64] (0004_updates)        16.6346 (1.11)   
test_freq_shift_gpu[0.3-1.0-256-float64] (0004_updates)         16.6695 (1.11)   
test_freq_shift_gpu[0.3--3.0-256-float64] (0004_updates)        16.6730 (1.11)   
test_freq_shift_gpu[0.3-2.0-256-float64] (0004_updates)         16.7105 (1.12)   
test_freq_shift_gpu[0.3-3.0-256-float64] (0004_updates)         16.7335 (1.12)   
test_freq_shift_gpu[0.3-1.0-256-complex128] (0004_updates)      16.8149 (1.12)   
test_freq_shift_gpu[0.3--5.0-256-complex128] (0004_updates)     16.9078 (1.13)   
test_freq_shift_gpu[0.3--3.0-256-complex128] (0004_updates)     16.9696 (1.13)   
test_freq_shift_gpu[0.3--2.0-256-complex128] (0004_updates)     16.9932 (1.13)   
test_freq_shift_gpu[0.3-3.0-256-complex128] (0004_updates)      16.9993 (1.14)   
test_freq_shift_gpu[0.3-2.0-256-complex128] (0004_updates)      17.0141 (1.14)   
test_freq_shift_gpu[0.3--4.0-256-complex128] (0004_updates)     17.0351 (1.14)   
test_freq_shift_gpu[0.3-4.0-256-complex128] (0004_updates)      17.0872 (1.14)   
test_freq_shift_gpu[0.3--1.0-256-complex128] (0004_updates)     17.1756 (1.15)   
test_freq_shift_gpu[0.3-0.0-256-float64] (0005_branch-)         52.3198 (3.49)   
test_freq_shift_gpu[0.3--5.0-256-float64] (0005_branch-)        52.4405 (3.50)   
test_freq_shift_gpu[0.3--4.0-256-float64] (0005_branch-)        52.4583 (3.50)   
test_freq_shift_gpu[0.3--3.0-256-float64] (0005_branch-)        52.5530 (3.51)   
test_freq_shift_gpu[0.3-3.0-256-float64] (0005_branch-)         52.6829 (3.52)   
test_freq_shift_gpu[0.3--2.0-256-float64] (0005_branch-)        52.8396 (3.53)   
test_freq_shift_gpu[0.3-2.0-256-float64] (0005_branch-)         52.8623 (3.53)   
test_freq_shift_gpu[0.3--4.0-256-complex128] (0005_branch-)     52.9230 (3.53)   
test_freq_shift_gpu[0.3-4.0-256-float64] (0005_branch-)         52.9261 (3.53)   
test_freq_shift_gpu[0.3-1.0-256-float64] (0005_branch-)         53.0048 (3.54)   
test_freq_shift_gpu[0.3--1.0-256-float64] (0005_branch-)        53.0116 (3.54)   
test_freq_shift_gpu[0.3-4.0-256-complex128] (0005_branch-)      53.0403 (3.54)   
test_freq_shift_gpu[0.3--3.0-256-complex128] (0005_branch-)     53.1384 (3.55)   
test_freq_shift_gpu[0.3-3.0-256-complex128] (0005_branch-)      53.3254 (3.56)   
test_freq_shift_gpu[0.3-0.0-256-complex128] (0005_branch-)      53.4006 (3.57)   
test_freq_shift_gpu[0.3-2.0-256-complex128] (0005_branch-)      53.4921 (3.57)   
test_freq_shift_gpu[0.3-1.0-256-complex128] (0005_branch-)      53.5231 (3.57)   
test_freq_shift_gpu[0.3--1.0-256-complex128] (0005_branch-)     53.5735 (3.58)   
test_freq_shift_gpu[0.3--5.0-256-complex128] (0005_branch-)     53.5927 (3.58)   
test_freq_shift_gpu[0.3--2.0-256-complex128] (0005_branch-)     54.0957 (3.61)   
---------------------------------------------------------------------------------


----------------- benchmark 'GaussPulse': 4 tests ------------------
Name (time in us)                                     Mean          
--------------------------------------------------------------------
test_gausspulse_gpu[0.75-16384] (0004_updates)     28.4977 (1.0)    
test_gausspulse_gpu[5-16384] (0004_updates)        28.9255 (1.02)   
test_gausspulse_gpu[0.75-16384] (0005_branch-)     29.0761 (1.02)   
test_gausspulse_gpu[5-16384] (0005_branch-)        29.3997 (1.03)   
--------------------------------------------------------------------


----------------- benchmark 'GaussSpline': 2 tests ----------------
Name (time in us)                                    Mean          
-------------------------------------------------------------------
test_gauss_spline_gpu[1-65536] (0005_branch-)     27.2676 (1.0)    
test_gauss_spline_gpu[1-65536] (0004_updates)     28.2749 (1.04)   
-------------------------------------------------------------------


---------------- benchmark 'Gaussian': 4 tests ----------------
Name (time in us)                                Mean          
---------------------------------------------------------------
test_gaussian_gpu[3-32768] (0004_updates)     16.6387 (1.0)    
test_gaussian_gpu[7-32768] (0004_updates)     16.7228 (1.01)   
test_gaussian_gpu[3-32768] (0005_branch-)     17.4171 (1.05)   
test_gaussian_gpu[7-32768] (0005_branch-)     17.4423 (1.05)   
---------------------------------------------------------------


--------------- benchmark 'GeneralCosine': 2 tests ---------------
Name (time in ms)                                   Mean          
------------------------------------------------------------------
test_general_cosine_gpu[32768] (0004_updates)     1.2663 (1.0)    
test_general_cosine_gpu[32768] (0005_branch-)     1.2831 (1.01)   
------------------------------------------------------------------


------------------- benchmark 'GeneralGaussian': 8 tests -------------------
Name (time in us)                                             Mean          
----------------------------------------------------------------------------
test_general_gaussian_gpu[7-1.5-32768] (0004_updates)      30.2709 (1.0)    
test_general_gaussian_gpu[3-1.5-32768] (0004_updates)      30.2883 (1.00)   
test_general_gaussian_gpu[7-0.75-32768] (0004_updates)     30.4432 (1.01)   
test_general_gaussian_gpu[7-1.5-32768] (0005_branch-)      30.9246 (1.02)   
test_general_gaussian_gpu[3-0.75-32768] (0004_updates)     30.9494 (1.02)   
test_general_gaussian_gpu[7-0.75-32768] (0005_branch-)     31.1184 (1.03)   
test_general_gaussian_gpu[3-1.5-32768] (0005_branch-)      31.1534 (1.03)   
test_general_gaussian_gpu[3-0.75-32768] (0005_branch-)     31.7321 (1.05)   
----------------------------------------------------------------------------


------------------ benchmark 'GeneralHamming': 4 tests ------------------
Name (time in us)                                          Mean          
-------------------------------------------------------------------------
test_general_hamming_gpu[0.25-32768] (0004_updates)     49.7750 (1.0)    
test_general_hamming_gpu[0.5-32768] (0004_updates)      52.0314 (1.05)   
test_general_hamming_gpu[0.25-32768] (0005_branch-)     52.8641 (1.06)   
test_general_hamming_gpu[0.5-32768] (0005_branch-)      53.6020 (1.08)   
-------------------------------------------------------------------------


-------------------- benchmark 'GetWindow': 6 tests -------------------
Name (time in us)                                        Mean          
-----------------------------------------------------------------------
test_get_window_gpu[32768-triang] (0004_updates)      17.0104 (1.0)    
test_get_window_gpu[32768-boxcar] (0004_updates)      18.2042 (1.07)   
test_get_window_gpu[32768-triang] (0005_branch-)      18.7202 (1.10)   
test_get_window_gpu[32768-boxcar] (0005_branch-)      18.8602 (1.11)   
test_get_window_gpu[32768-nuttall] (0004_updates)     65.7180 (3.86)   
test_get_window_gpu[32768-nuttall] (0005_branch-)     66.1057 (3.89)   
-----------------------------------------------------------------------


--------------- benchmark 'Hamming': 2 tests ---------------
Name (time in us)                             Mean          
------------------------------------------------------------
test_hamming_gpu[32768] (0004_updates)     15.8773 (1.0)    
test_hamming_gpu[32768] (0005_branch-)     17.8413 (1.12)   
------------------------------------------------------------


----------------- benchmark 'Hilbert': 4 tests ----------------
Name (time in us)                                Mean          
---------------------------------------------------------------
test_hilbert_gpu[2-256] (0004_updates)        99.5911 (1.0)    
test_hilbert_gpu[2-256] (0005_branch-)       128.9426 (1.29)   
test_hilbert_gpu[1-32768] (0004_updates)     144.0002 (1.45)   
test_hilbert_gpu[1-32768] (0005_branch-)     156.7514 (1.57)   
---------------------------------------------------------------


---------- benchmark 'Hilbert2': 1 tests -----------
Name (time in us)                               Mean
----------------------------------------------------
test_hilbert2_gpu[2-256] (0004_updates)     159.2338
----------------------------------------------------


----------------- benchmark 'InverseComplexCepstrum': 2 tests ------------------
Name (time in us)                                                 Mean          
--------------------------------------------------------------------------------
test_inverse_complex_cepstrum_gpu[256-1024] (0004_updates)     86.6944 (1.0)    
test_inverse_complex_cepstrum_gpu[123-1024] (0004_updates)     87.8373 (1.01)   
--------------------------------------------------------------------------------


----------------- benchmark 'Kaiser': 4 tests ------------------
Name (time in us)                                 Mean          
----------------------------------------------------------------
test_kaiser_gpu[0.5-32768] (0004_updates)      21.8793 (1.0)    
test_kaiser_gpu[0.25-32768] (0004_updates)     22.2693 (1.02)   
test_kaiser_gpu[0.5-32768] (0005_branch-)      22.5218 (1.03)   
test_kaiser_gpu[0.25-32768] (0005_branch-)     23.8247 (1.09)   
----------------------------------------------------------------


----------------- benchmark 'KaiserAtten': 18 tests -----------------
Name (time in us)                                      Mean          
---------------------------------------------------------------------
test_kaiser_atten_gpu[0.0375-5] (0004_updates)       1.3759 (1.0)    
test_kaiser_atten_gpu[2.4-100] (0004_updates)        1.3795 (1.00)   
test_kaiser_atten_gpu[0.0375-100] (0004_updates)     1.3797 (1.00)   
test_kaiser_atten_gpu[2.4-5] (0004_updates)          1.3907 (1.01)   
test_kaiser_atten_gpu[0.0375-25] (0004_updates)      1.3922 (1.01)   
test_kaiser_atten_gpu[2.4-25] (0004_updates)         1.3936 (1.01)   
test_kaiser_atten_gpu[0.01-25] (0004_updates)        1.3960 (1.01)   
test_kaiser_atten_gpu[0.01-5] (0004_updates)         1.4006 (1.02)   
test_kaiser_atten_gpu[0.01-100] (0004_updates)       1.4011 (1.02)   
test_kaiser_atten_gpu[0.0375-100] (0005_branch-)     1.4164 (1.03)   
test_kaiser_atten_gpu[0.0375-5] (0005_branch-)       1.4183 (1.03)   
test_kaiser_atten_gpu[0.01-25] (0005_branch-)        1.4232 (1.03)   
test_kaiser_atten_gpu[0.01-5] (0005_branch-)         1.4274 (1.04)   
test_kaiser_atten_gpu[2.4-25] (0005_branch-)         1.4276 (1.04)   
test_kaiser_atten_gpu[2.4-100] (0005_branch-)        1.4290 (1.04)   
test_kaiser_atten_gpu[0.01-100] (0005_branch-)       1.4327 (1.04)   
test_kaiser_atten_gpu[2.4-5] (0005_branch-)          1.4353 (1.04)   
test_kaiser_atten_gpu[0.0375-25] (0005_branch-)      1.4380 (1.05)   
---------------------------------------------------------------------


-------------- benchmark 'KaiserBeta': 6 tests --------------
Name (time in us)                              Mean          
-------------------------------------------------------------
test_kaiser_beta_gpu[5] (0004_updates)       1.2753 (1.0)    
test_kaiser_beta_gpu[100] (0004_updates)     1.3353 (1.05)   
test_kaiser_beta_gpu[5] (0005_branch-)       1.3675 (1.07)   
test_kaiser_beta_gpu[100] (0005_branch-)     1.4185 (1.11)   
test_kaiser_beta_gpu[25] (0004_updates)      1.4426 (1.13)   
test_kaiser_beta_gpu[25] (0005_branch-)      1.5126 (1.19)   
-------------------------------------------------------------


------------------------- benchmark 'LombScargle': 16 tests -------------------------
Name (time in us)                                                      Mean          
-------------------------------------------------------------------------------------
test_lombscargle_gpu[False-False-65536-1024] (0004_updates)        921.0315 (1.0)    
test_lombscargle_gpu[False-False-65536-1024] (0005_branch-)        931.4745 (1.01)   
test_lombscargle_gpu[True-False-65536-1024] (0004_updates)         995.4225 (1.08)   
test_lombscargle_gpu[False-True-65536-1024] (0005_branch-)       1,035.1708 (1.12)   
test_lombscargle_gpu[True-False-65536-1024] (0005_branch-)       1,065.5929 (1.16)   
test_lombscargle_gpu[False-True-65536-1024] (0004_updates)       1,075.3241 (1.17)   
test_lombscargle_gpu[True-True-65536-1024] (0004_updates)        1,090.2164 (1.18)   
test_lombscargle_gpu[True-True-65536-1024] (0005_branch-)        1,248.5989 (1.36)   
test_lombscargle_gpu[True-False-262144-1024] (0004_updates)      3,591.7437 (3.90)   
test_lombscargle_gpu[False-True-262144-1024] (0004_updates)      3,700.3981 (4.02)   
test_lombscargle_gpu[True-False-262144-1024] (0005_branch-)      3,747.1127 (4.07)   
test_lombscargle_gpu[True-True-262144-1024] (0004_updates)       3,831.1901 (4.16)   
test_lombscargle_gpu[False-False-262144-1024] (0004_updates)     4,013.9554 (4.36)   
test_lombscargle_gpu[False-True-262144-1024] (0005_branch-)      4,087.8167 (4.44)   
test_lombscargle_gpu[False-False-262144-1024] (0005_branch-)     4,162.7265 (4.52)   
test_lombscargle_gpu[True-True-262144-1024] (0005_branch-)       4,233.6392 (4.60)   
-------------------------------------------------------------------------------------


------------------ benchmark 'MinimumPhase': 4 tests ------------------
Name (time in us)                                        Mean          
-----------------------------------------------------------------------
test_minimum_phase_gpu[256-256] (0004_updates)       169.9252 (1.0)    
test_minimum_phase_gpu[256-16384] (0004_updates)     188.3045 (1.11)   
test_minimum_phase_gpu[123-16384] (0004_updates)     193.2694 (1.14)   
test_minimum_phase_gpu[123-256] (0004_updates)       196.1153 (1.15)   
-----------------------------------------------------------------------


--------------- benchmark 'Morlet': 2 tests ---------------
Name (time in us)                            Mean          
-----------------------------------------------------------
test_morlet_gpu[16384] (0004_updates)     19.7552 (1.0)    
test_morlet_gpu[16384] (0005_branch-)     27.8710 (1.41)   
-----------------------------------------------------------


--------------- benchmark 'Nuttall': 2 tests ---------------
Name (time in us)                             Mean          
------------------------------------------------------------
test_nuttall_gpu[32768] (0004_updates)     55.9443 (1.0)    
test_nuttall_gpu[32768] (0005_branch-)     58.5968 (1.05)   
------------------------------------------------------------


--------------- benchmark 'Parzen': 3 tests ---------------
Name (time in us)                            Mean          
-----------------------------------------------------------
test_parzen_gpu[32767] (0004_updates)     16.3731 (1.0)    
test_parzen_gpu[32768] (0004_updates)     17.3191 (1.06)   
test_parzen_gpu[32768] (0005_branch-)     17.3798 (1.06)   
-----------------------------------------------------------


---------------------------------- benchmark 'Periodogram': 32 tests ----------------------------------
Name (time in us)                                                                        Mean          
-------------------------------------------------------------------------------------------------------
test_periodogram_gpu[spectrum-flattop-1000000.0-16384-float64] (0004_updates)        419.6845 (1.0)    
test_periodogram_gpu[spectrum-nuttall-1000000.0-16384-float64] (0004_updates)        420.0483 (1.00)   
test_periodogram_gpu[spectrum-nuttall-1.0-16384-float64] (0004_updates)              421.2155 (1.00)   
test_periodogram_gpu[spectrum-flattop-1.0-16384-float64] (0004_updates)              423.7665 (1.01)   
test_periodogram_gpu[density-nuttall-1.0-16384-float64] (0004_updates)               424.6190 (1.01)   
test_periodogram_gpu[spectrum-flattop-1.0-16384-float64] (0005_branch-)              424.7606 (1.01)   
test_periodogram_gpu[spectrum-nuttall-1.0-16384-float64] (0005_branch-)              424.7733 (1.01)   
test_periodogram_gpu[spectrum-flattop-1000000.0-16384-float64] (0005_branch-)        425.2234 (1.01)   
test_periodogram_gpu[density-flattop-1.0-16384-float64] (0004_updates)               425.8457 (1.01)   
test_periodogram_gpu[density-nuttall-1000000.0-16384-float64] (0004_updates)         427.2919 (1.02)   
test_periodogram_gpu[density-flattop-1000000.0-16384-float64] (0004_updates)         427.6556 (1.02)   
test_periodogram_gpu[density-flattop-1.0-16384-float64] (0005_branch-)               430.4081 (1.03)   
test_periodogram_gpu[density-nuttall-1.0-16384-float64] (0005_branch-)               431.0347 (1.03)   
test_periodogram_gpu[spectrum-nuttall-1000000.0-16384-float64] (0005_branch-)        431.7973 (1.03)   
test_periodogram_gpu[density-flattop-1000000.0-16384-float64] (0005_branch-)         433.0106 (1.03)   
test_periodogram_gpu[density-nuttall-1000000.0-16384-float64] (0005_branch-)         436.1358 (1.04)   
test_periodogram_gpu[spectrum-nuttall-1000000.0-16384-complex128] (0004_updates)     635.1603 (1.51)   
test_periodogram_gpu[spectrum-nuttall-1.0-16384-complex128] (0004_updates)           635.5525 (1.51)   
test_periodogram_gpu[spectrum-flattop-1.0-16384-complex128] (0004_updates)           636.9335 (1.52)   
test_periodogram_gpu[spectrum-flattop-1000000.0-16384-complex128] (0004_updates)     637.4375 (1.52)   
test_periodogram_gpu[spectrum-nuttall-1000000.0-16384-complex128] (0005_branch-)     638.9030 (1.52)   
test_periodogram_gpu[spectrum-nuttall-1.0-16384-complex128] (0005_branch-)           640.1748 (1.53)   
test_periodogram_gpu[spectrum-flattop-1000000.0-16384-complex128] (0005_branch-)     640.5334 (1.53)   
test_periodogram_gpu[spectrum-flattop-1.0-16384-complex128] (0005_branch-)           641.8317 (1.53)   
test_periodogram_gpu[density-nuttall-1.0-16384-complex128] (0004_updates)            644.3246 (1.54)   
test_periodogram_gpu[density-flattop-1.0-16384-complex128] (0005_branch-)            644.5762 (1.54)   
test_periodogram_gpu[density-nuttall-1000000.0-16384-complex128] (0005_branch-)      644.7801 (1.54)   
test_periodogram_gpu[density-nuttall-1000000.0-16384-complex128] (0004_updates)      645.1077 (1.54)   
test_periodogram_gpu[density-flattop-1.0-16384-complex128] (0004_updates)            645.7488 (1.54)   
test_periodogram_gpu[density-nuttall-1.0-16384-complex128] (0005_branch-)            645.9917 (1.54)   
test_periodogram_gpu[density-flattop-1000000.0-16384-complex128] (0004_updates)      646.6338 (1.54)   
test_periodogram_gpu[density-flattop-1000000.0-16384-complex128] (0005_branch-)      648.3478 (1.54)   
-------------------------------------------------------------------------------------------------------


--------------- benchmark 'Qmf': 2 tests ---------------
Name (time in us)                         Mean          
--------------------------------------------------------
test_qmf_gpu[16384] (0004_updates)     10.8242 (1.0)    
test_qmf_gpu[16384] (0005_branch-)     12.3952 (1.15)   
--------------------------------------------------------


--------------- benchmark 'Quadratic': 2 tests ---------------
Name (time in us)                               Mean          
--------------------------------------------------------------
test_quadratic_gpu[65536] (0004_updates)     15.0789 (1.0)    
test_quadratic_gpu[65536] (0005_branch-)     16.0059 (1.06)   
--------------------------------------------------------------


------------------ benchmark 'RealCepstrum': 8 tests ------------------
Name (time in us)                                        Mean          
-----------------------------------------------------------------------
test_real_cepstrum_gpu[256-256] (0004_updates)        83.4747 (1.0)    
test_real_cepstrum_gpu[256-16384] (0004_updates)     100.2308 (1.20)   
test_real_cepstrum_gpu[123-16384] (0004_updates)     102.9805 (1.23)   
test_real_cepstrum_gpu[123-256] (0004_updates)       105.4617 (1.26)   
test_real_cepstrum_gpu[256-256] (0005_branch-)       114.7432 (1.37)   
test_real_cepstrum_gpu[256-16384] (0005_branch-)     126.8742 (1.52)   
test_real_cepstrum_gpu[123-256] (0005_branch-)       136.0628 (1.63)   
test_real_cepstrum_gpu[123-16384] (0005_branch-)     139.4896 (1.67)   
-----------------------------------------------------------------------


---------------------- benchmark 'Resample': 4 tests -----------------------
Name (time in us)                                             Mean          
----------------------------------------------------------------------------
test_resample_gpu[window0-65536-16384] (0004_updates)     226.0914 (1.0)    
test_resample_gpu[window0-4096-16384] (0004_updates)      231.5319 (1.02)   
test_resample_gpu[window0-65536-16384] (0005_branch-)     266.5696 (1.18)   
test_resample_gpu[window0-4096-16384] (0005_branch-)      274.3969 (1.21)   
----------------------------------------------------------------------------


---------------------- benchmark 'ResamplePoly': 18 tests ---------------------
Name (time in us)                                                Mean          
-------------------------------------------------------------------------------
test_resample_poly_gpu[window0-2-2-16384] (0004_updates)      18.0865 (1.0)    
test_resample_poly_gpu[window0-2-2-16384] (0005_branch-)      18.4871 (1.02)   
test_resample_poly_gpu[window0-1-2-16384] (0004_updates)     606.4266 (33.53)  
test_resample_poly_gpu[window0-9-7-16384] (0004_updates)     607.2294 (33.57)  
test_resample_poly_gpu[window0-1-3-16384] (0004_updates)     609.9951 (33.73)  
test_resample_poly_gpu[window0-2-7-16384] (0004_updates)     613.5200 (33.92)  
test_resample_poly_gpu[window0-2-3-16384] (0004_updates)     615.1580 (34.01)  
test_resample_poly_gpu[window0-1-7-16384] (0004_updates)     621.8735 (34.38)  
test_resample_poly_gpu[window0-9-3-16384] (0004_updates)     624.4177 (34.52)  
test_resample_poly_gpu[window0-9-2-16384] (0004_updates)     639.6979 (35.37)  
test_resample_poly_gpu[window0-1-2-16384] (0005_branch-)     824.2699 (45.57)  
test_resample_poly_gpu[window0-1-3-16384] (0005_branch-)     828.8350 (45.83)  
test_resample_poly_gpu[window0-2-3-16384] (0005_branch-)     832.6161 (46.04)  
test_resample_poly_gpu[window0-1-7-16384] (0005_branch-)     833.8360 (46.10)  
test_resample_poly_gpu[window0-2-7-16384] (0005_branch-)     835.7921 (46.21)  
test_resample_poly_gpu[window0-9-3-16384] (0005_branch-)     837.1205 (46.28)  
test_resample_poly_gpu[window0-9-7-16384] (0005_branch-)     847.6794 (46.87)  
test_resample_poly_gpu[window0-9-2-16384] (0005_branch-)     872.4615 (48.24)  
-------------------------------------------------------------------------------


----------------- benchmark 'Ricker': 4 tests ------------------
Name (time in us)                                 Mean          
----------------------------------------------------------------
test_ricker_gpu[10-16384] (0004_updates)       15.5962 (1.0)    
test_ricker_gpu[1000-16384] (0004_updates)     15.6274 (1.00)   
test_ricker_gpu[10-16384] (0005_branch-)       20.3913 (1.31)   
test_ricker_gpu[1000-16384] (0005_branch-)     20.8291 (1.34)   
----------------------------------------------------------------


----------------------- benchmark 'SOSFilt': 24 tests -----------------------
Name (time in ms)                                              Mean          
-----------------------------------------------------------------------------
test_sosfilt_gpu[float64-1-32768-32] (0004_updates)          8.3189 (1.0)    
test_sosfilt_gpu[float64-10-32768-32] (0005_branch-)         8.3239 (1.00)   
test_sosfilt_gpu[float64-2-32768-32] (0004_updates)          8.3266 (1.00)   
test_sosfilt_gpu[float64-1-32768-32] (0005_branch-)          8.3284 (1.00)   
test_sosfilt_gpu[float64-10-32768-32] (0004_updates)         8.3293 (1.00)   
test_sosfilt_gpu[float64-2-32768-32] (0005_branch-)          8.3432 (1.00)   
test_sosfilt_gpu[float64-1-32768-64] (0005_branch-)          8.5644 (1.03)   
test_sosfilt_gpu[float64-1-32768-64] (0004_updates)          8.5646 (1.03)   
test_sosfilt_gpu[float64-10-32768-64] (0005_branch-)         8.5786 (1.03)   
test_sosfilt_gpu[float64-10-32768-64] (0004_updates)         8.5813 (1.03)   
test_sosfilt_gpu[float64-2-32768-64] (0004_updates)          8.5828 (1.03)   
test_sosfilt_gpu[float64-2-32768-64] (0005_branch-)          8.5870 (1.03)   
test_sosfilt_gpu[float64-1-1048576-32] (0004_updates)      277.7020 (33.38)  
test_sosfilt_gpu[float64-2-1048576-32] (0004_updates)      277.7039 (33.38)  
test_sosfilt_gpu[float64-1-1048576-32] (0005_branch-)      277.7459 (33.39)  
test_sosfilt_gpu[float64-2-1048576-32] (0005_branch-)      277.7493 (33.39)  
test_sosfilt_gpu[float64-10-1048576-32] (0004_updates)     277.9877 (33.42)  
test_sosfilt_gpu[float64-10-1048576-32] (0005_branch-)     278.0150 (33.42)  
test_sosfilt_gpu[float64-1-1048576-64] (0004_updates)      285.0072 (34.26)  
test_sosfilt_gpu[float64-2-1048576-64] (0004_updates)      285.0292 (34.26)  
test_sosfilt_gpu[float64-1-1048576-64] (0005_branch-)      285.0444 (34.26)  
test_sosfilt_gpu[float64-2-1048576-64] (0005_branch-)      285.0750 (34.27)  
test_sosfilt_gpu[float64-10-1048576-64] (0004_updates)     285.2873 (34.29)  
test_sosfilt_gpu[float64-10-1048576-64] (0005_branch-)     285.3245 (34.30)  
-----------------------------------------------------------------------------


---------------------------- benchmark 'STFT': 16 tests ----------------------------
Name (time in us)                                                     Mean          
------------------------------------------------------------------------------------
test_stft_gpu[1024-1.0-16384-float64] (0004_updates)              348.6882 (1.0)    
test_stft_gpu[1024-1.0-16384-float64] (0005_branch-)              349.7215 (1.00)   
test_stft_gpu[1024-1000000.0-16384-float64] (0004_updates)        349.7221 (1.00)   
test_stft_gpu[2048-1000000.0-16384-float64] (0004_updates)        350.4510 (1.01)   
test_stft_gpu[2048-1000000.0-16384-float64] (0005_branch-)        352.0547 (1.01)   
test_stft_gpu[2048-1.0-16384-float64] (0004_updates)              352.3533 (1.01)   
test_stft_gpu[2048-1.0-16384-float64] (0005_branch-)              353.3257 (1.01)   
test_stft_gpu[1024-1000000.0-16384-float64] (0005_branch-)        354.2423 (1.02)   
test_stft_gpu[2048-1000000.0-16384-complex128] (0004_updates)     394.7264 (1.13)   
test_stft_gpu[1024-1000000.0-16384-complex128] (0004_updates)     395.3608 (1.13)   
test_stft_gpu[2048-1.0-16384-complex128] (0004_updates)           395.9015 (1.14)   
test_stft_gpu[1024-1.0-16384-complex128] (0004_updates)           396.6526 (1.14)   
test_stft_gpu[1024-1000000.0-16384-complex128] (0005_branch-)     405.2387 (1.16)   
test_stft_gpu[1024-1.0-16384-complex128] (0005_branch-)           406.5403 (1.17)   
test_stft_gpu[2048-1000000.0-16384-complex128] (0005_branch-)     407.6533 (1.17)   
test_stft_gpu[2048-1.0-16384-complex128] (0005_branch-)           414.5012 (1.19)   
------------------------------------------------------------------------------------


---------------------------- benchmark 'Spectrogram': 16 tests ----------------------------
Name (time in us)                                                            Mean          
-------------------------------------------------------------------------------------------
test_spectrogram_gpu[2048-1000000.0-16384-float64] (0004_updates)        319.3857 (1.0)    
test_spectrogram_gpu[2048-1.0-16384-float64] (0004_updates)              320.1697 (1.00)   
test_spectrogram_gpu[1024-1000000.0-16384-float64] (0004_updates)        322.4024 (1.01)   
test_spectrogram_gpu[1024-1.0-16384-float64] (0004_updates)              324.5655 (1.02)   
test_spectrogram_gpu[2048-1000000.0-16384-float64] (0005_branch-)        325.4783 (1.02)   
test_spectrogram_gpu[2048-1.0-16384-float64] (0005_branch-)              326.9566 (1.02)   
test_spectrogram_gpu[1024-1000000.0-16384-float64] (0005_branch-)        330.3982 (1.03)   
test_spectrogram_gpu[1024-1.0-16384-float64] (0005_branch-)              333.1907 (1.04)   
test_spectrogram_gpu[1024-1.0-16384-complex128] (0004_updates)           345.1062 (1.08)   
test_spectrogram_gpu[1024-1000000.0-16384-complex128] (0004_updates)     345.1483 (1.08)   
test_spectrogram_gpu[2048-1.0-16384-complex128] (0004_updates)           347.2771 (1.09)   
test_spectrogram_gpu[2048-1000000.0-16384-complex128] (0004_updates)     348.0628 (1.09)   
test_spectrogram_gpu[1024-1000000.0-16384-complex128] (0005_branch-)     349.6415 (1.09)   
test_spectrogram_gpu[1024-1.0-16384-complex128] (0005_branch-)           350.1164 (1.10)   
test_spectrogram_gpu[2048-1.0-16384-complex128] (0005_branch-)           356.3489 (1.12)   
test_spectrogram_gpu[2048-1000000.0-16384-complex128] (0005_branch-)     363.5744 (1.14)   
-------------------------------------------------------------------------------------------


----------------- benchmark 'Square': 4 tests ------------------
Name (time in us)                                 Mean          
----------------------------------------------------------------
test_square_gpu[0.5-16384] (0004_updates)      26.7124 (1.0)    
test_square_gpu[0.25-16384] (0004_updates)     26.8315 (1.00)   
test_square_gpu[0.5-16384] (0005_branch-)      33.8961 (1.27)   
test_square_gpu[0.25-16384] (0005_branch-)     33.9864 (1.27)   
----------------------------------------------------------------


------------------- benchmark 'TestArgrelmax': 24 tests -------------------
Name (time in us)                                            Mean          
---------------------------------------------------------------------------
test_argrelmax_gpu[wrap-1--1-1-32768] (0004_updates)      99.8563 (1.0)    
test_argrelmax_gpu[clip-2--1-1-32768] (0004_updates)      99.8967 (1.00)   
test_argrelmax_gpu[clip-1--1-1-32768] (0004_updates)     100.9340 (1.01)   
test_argrelmax_gpu[wrap-2--1-1-32768] (0004_updates)     101.1204 (1.01)   
test_argrelmax_gpu[wrap-2--1-2-256] (0004_updates)       104.0368 (1.04)   
test_argrelmax_gpu[wrap-2--1-1-32768] (0005_branch-)     104.1482 (1.04)   
test_argrelmax_gpu[clip-2--1-1-32768] (0005_branch-)     104.4796 (1.05)   
test_argrelmax_gpu[wrap-1--1-2-256] (0004_updates)       104.5792 (1.05)   
test_argrelmax_gpu[clip-1--1-1-32768] (0005_branch-)     104.7046 (1.05)   
test_argrelmax_gpu[clip-2--1-2-256] (0004_updates)       104.7206 (1.05)   
test_argrelmax_gpu[wrap-1--1-1-32768] (0005_branch-)     104.7338 (1.05)   
test_argrelmax_gpu[clip-1--1-2-256] (0004_updates)       105.0495 (1.05)   
test_argrelmax_gpu[clip-1--1-2-256] (0005_branch-)       109.2574 (1.09)   
test_argrelmax_gpu[wrap-2--1-2-256] (0005_branch-)       109.4445 (1.10)   
test_argrelmax_gpu[wrap-1--1-2-256] (0005_branch-)       109.4492 (1.10)   
test_argrelmax_gpu[clip-2--1-2-256] (0005_branch-)       109.6248 (1.10)   
test_argrelmax_gpu[wrap-1--1-3-32] (0004_updates)        270.6945 (2.71)   
test_argrelmax_gpu[wrap-1--1-3-32] (0005_branch-)        286.4023 (2.87)   
test_argrelmax_gpu[clip-1--1-3-32] (0004_updates)        322.1066 (3.23)   
test_argrelmax_gpu[clip-1--1-3-32] (0005_branch-)        336.9644 (3.37)   
test_argrelmax_gpu[wrap-2--1-3-32] (0004_updates)        405.9938 (4.07)   
test_argrelmax_gpu[wrap-2--1-3-32] (0005_branch-)        418.0348 (4.19)   
test_argrelmax_gpu[clip-2--1-3-32] (0004_updates)        503.3482 (5.04)   
test_argrelmax_gpu[clip-2--1-3-32] (0005_branch-)        522.6791 (5.23)   
---------------------------------------------------------------------------


--------------- benchmark 'Triang': 2 tests ---------------
Name (time in us)                            Mean          
-----------------------------------------------------------
test_triang_gpu[32768] (0004_updates)     13.2609 (1.0)    
test_triang_gpu[32768] (0005_branch-)     14.9778 (1.13)   
-----------------------------------------------------------


------------------ benchmark 'Tukey': 4 tests -----------------
Name (time in us)                                Mean          
---------------------------------------------------------------
test_tukey_gpu[0.5-32768] (0004_updates)      17.7753 (1.0)    
test_tukey_gpu[0.25-32768] (0004_updates)     18.3326 (1.03)   
test_tukey_gpu[0.25-32768] (0005_branch-)     23.3924 (1.32)   
test_tukey_gpu[0.5-32768] (0005_branch-)      23.5339 (1.32)   
---------------------------------------------------------------


------------------ benchmark 'UnitImpulse': 2 tests -----------------
Name (time in us)                                      Mean          
---------------------------------------------------------------------
test_unit_impulse_gpu[mid-16384] (0004_updates)     22.0059 (1.0)    
test_unit_impulse_gpu[mid-16384] (0005_branch-)     23.4531 (1.07)   
---------------------------------------------------------------------


------------------- benchmark 'UpFirDn': 72 tests --------------------
Name (time in us)                                       Mean          
----------------------------------------------------------------------
test_upfirdn_gpu[0-9-7-1-16384] (0004_updates)      200.7410 (1.0)    
test_upfirdn_gpu[-1-2-7-1-16384] (0004_updates)     200.7895 (1.00)   
test_upfirdn_gpu[0-2-7-1-16384] (0004_updates)      201.1609 (1.00)   
test_upfirdn_gpu[-1-9-3-2-256] (0004_updates)       201.6581 (1.00)   
test_upfirdn_gpu[-1-9-7-2-256] (0004_updates)       201.8256 (1.01)   
test_upfirdn_gpu[0-2-3-1-16384] (0004_updates)      202.1470 (1.01)   
test_upfirdn_gpu[0-1-3-1-16384] (0004_updates)      202.2788 (1.01)   
test_upfirdn_gpu[-1-2-3-1-16384] (0004_updates)     202.2796 (1.01)   
test_upfirdn_gpu[0-9-3-1-16384] (0004_updates)      202.3289 (1.01)   
test_upfirdn_gpu[0-1-7-1-16384] (0004_updates)      202.3522 (1.01)   
test_upfirdn_gpu[-1-9-7-1-16384] (0004_updates)     202.4565 (1.01)   
test_upfirdn_gpu[-1-9-3-1-16384] (0004_updates)     202.6086 (1.01)   
test_upfirdn_gpu[-1-1-7-1-16384] (0004_updates)     202.6880 (1.01)   
test_upfirdn_gpu[-1-2-3-2-256] (0004_updates)       202.8743 (1.01)   
test_upfirdn_gpu[0-9-7-2-256] (0004_updates)        203.1670 (1.01)   
test_upfirdn_gpu[0-9-3-2-256] (0004_updates)        203.2244 (1.01)   
test_upfirdn_gpu[0-1-3-2-256] (0004_updates)        203.3249 (1.01)   
test_upfirdn_gpu[0-2-7-2-256] (0004_updates)        203.8907 (1.02)   
test_upfirdn_gpu[-1-1-3-1-16384] (0004_updates)     204.0206 (1.02)   
test_upfirdn_gpu[-1-1-3-2-256] (0004_updates)       204.1716 (1.02)   
test_upfirdn_gpu[0-2-3-2-256] (0004_updates)        204.9088 (1.02)   
test_upfirdn_gpu[-1-2-7-2-256] (0004_updates)       205.6729 (1.02)   
test_upfirdn_gpu[0-1-3-1-16384] (0005_branch-)      207.8538 (1.04)   
test_upfirdn_gpu[-1-9-7-1-16384] (0005_branch-)     208.1739 (1.04)   
test_upfirdn_gpu[-1-1-3-1-16384] (0005_branch-)     208.4423 (1.04)   
test_upfirdn_gpu[0-1-7-1-16384] (0005_branch-)      208.5222 (1.04)   
test_upfirdn_gpu[-1-2-7-1-16384] (0005_branch-)     208.7680 (1.04)   
test_upfirdn_gpu[0-9-7-2-256] (0005_branch-)        208.8481 (1.04)   
test_upfirdn_gpu[0-2-3-2-256] (0005_branch-)        209.0433 (1.04)   
test_upfirdn_gpu[0-2-7-1-16384] (0005_branch-)      209.2506 (1.04)   
test_upfirdn_gpu[-1-2-3-1-16384] (0005_branch-)     209.3256 (1.04)   
test_upfirdn_gpu[0-9-3-2-256] (0005_branch-)        209.3278 (1.04)   
test_upfirdn_gpu[0-2-3-1-16384] (0005_branch-)      209.8397 (1.05)   
test_upfirdn_gpu[-1-9-3-2-256] (0005_branch-)       209.8478 (1.05)   
test_upfirdn_gpu[0-2-7-2-256] (0005_branch-)        210.2188 (1.05)   
test_upfirdn_gpu[-1-9-7-2-256] (0005_branch-)       210.5890 (1.05)   
test_upfirdn_gpu[0-1-7-2-256] (0004_updates)        210.7116 (1.05)   
test_upfirdn_gpu[-1-2-3-2-256] (0005_branch-)       210.7684 (1.05)   
test_upfirdn_gpu[0-1-3-2-256] (0005_branch-)        211.1961 (1.05)   
test_upfirdn_gpu[-1-1-7-1-16384] (0005_branch-)     211.6783 (1.05)   
test_upfirdn_gpu[-1-1-7-2-256] (0004_updates)       211.7534 (1.05)   
test_upfirdn_gpu[0-9-3-1-16384] (0005_branch-)      211.9375 (1.06)   
test_upfirdn_gpu[-1-9-3-1-16384] (0005_branch-)     212.5552 (1.06)   
test_upfirdn_gpu[-1-1-3-2-256] (0005_branch-)       212.7362 (1.06)   
test_upfirdn_gpu[0-9-7-1-16384] (0005_branch-)      212.9789 (1.06)   
test_upfirdn_gpu[-1-2-7-2-256] (0005_branch-)       213.8461 (1.07)   
test_upfirdn_gpu[0-2-2-1-16384] (0004_updates)      216.3090 (1.08)   
test_upfirdn_gpu[0-9-2-1-16384] (0004_updates)      216.9074 (1.08)   
test_upfirdn_gpu[-1-9-2-1-16384] (0004_updates)     217.2320 (1.08)   
test_upfirdn_gpu[0-1-7-2-256] (0005_branch-)        217.5610 (1.08)   
test_upfirdn_gpu[0-1-2-1-16384] (0004_updates)      217.8706 (1.09)   
test_upfirdn_gpu[-1-2-2-1-16384] (0004_updates)     218.4911 (1.09)   
test_upfirdn_gpu[0-1-2-2-256] (0004_updates)        218.7363 (1.09)   
test_upfirdn_gpu[-1-1-2-1-16384] (0004_updates)     218.9624 (1.09)   
test_upfirdn_gpu[-1-2-2-2-256] (0004_updates)       219.0186 (1.09)   
test_upfirdn_gpu[-1-9-2-2-256] (0004_updates)       219.3777 (1.09)   
test_upfirdn_gpu[0-2-2-2-256] (0004_updates)        219.4180 (1.09)   
test_upfirdn_gpu[-1-1-2-2-256] (0004_updates)       219.4305 (1.09)   
test_upfirdn_gpu[0-9-2-2-256] (0004_updates)        220.8575 (1.10)   
test_upfirdn_gpu[-1-1-7-2-256] (0005_branch-)       221.3077 (1.10)   
test_upfirdn_gpu[-1-1-2-1-16384] (0005_branch-)     225.8361 (1.13)   
test_upfirdn_gpu[0-2-2-2-256] (0005_branch-)        226.0363 (1.13)   
test_upfirdn_gpu[0-9-2-2-256] (0005_branch-)        226.3850 (1.13)   
test_upfirdn_gpu[0-1-2-1-16384] (0005_branch-)      226.5114 (1.13)   
test_upfirdn_gpu[0-9-2-1-16384] (0005_branch-)      227.1955 (1.13)   
test_upfirdn_gpu[-1-2-2-1-16384] (0005_branch-)     227.3958 (1.13)   
test_upfirdn_gpu[-1-9-2-1-16384] (0005_branch-)     227.7410 (1.13)   
test_upfirdn_gpu[-1-9-2-2-256] (0005_branch-)       228.2011 (1.14)   
test_upfirdn_gpu[-1-2-2-2-256] (0005_branch-)       228.6372 (1.14)   
test_upfirdn_gpu[0-1-2-2-256] (0005_branch-)        229.4443 (1.14)   
test_upfirdn_gpu[-1-1-2-2-256] (0005_branch-)       230.1536 (1.15)   
test_upfirdn_gpu[0-2-2-1-16384] (0005_branch-)      230.6118 (1.15)   
----------------------------------------------------------------------


------------------ benchmark 'Vectorstrength': 4 tests ------------------
Name (time in us)                                          Mean          
-------------------------------------------------------------------------
test_vectorstrength_gpu[16384-5] (0004_updates)        474.6993 (1.0)    
test_vectorstrength_gpu[16384-5] (0005_branch-)        478.8469 (1.01)   
test_vectorstrength_gpu[16384-0.75] (0004_updates)     492.3323 (1.04)   
test_vectorstrength_gpu[16384-0.75] (0005_branch-)     497.2639 (1.05)   
-------------------------------------------------------------------------


---------------------------- benchmark 'Welch': 16 tests ----------------------------
Name (time in us)                                                      Mean          
-------------------------------------------------------------------------------------
test_welch_gpu[2048-1.0-16384-float64] (0004_updates)              536.1166 (1.0)    
test_welch_gpu[1024-1.0-16384-float64] (0004_updates)              537.2353 (1.00)   
test_welch_gpu[1024-1000000.0-16384-float64] (0004_updates)        539.0315 (1.01)   
test_welch_gpu[2048-1000000.0-16384-float64] (0004_updates)        542.1727 (1.01)   
test_welch_gpu[1024-1000000.0-16384-float64] (0005_branch-)        547.9514 (1.02)   
test_welch_gpu[2048-1000000.0-16384-float64] (0005_branch-)        549.9385 (1.03)   
test_welch_gpu[2048-1.0-16384-float64] (0005_branch-)              550.0439 (1.03)   
test_welch_gpu[1024-1.0-16384-float64] (0005_branch-)              550.3895 (1.03)   
test_welch_gpu[2048-1000000.0-16384-complex128] (0004_updates)     577.3126 (1.08)   
test_welch_gpu[2048-1000000.0-16384-complex128] (0005_branch-)     579.1964 (1.08)   
test_welch_gpu[1024-1000000.0-16384-complex128] (0004_updates)     579.9484 (1.08)   
test_welch_gpu[2048-1.0-16384-complex128] (0004_updates)           580.0702 (1.08)   
test_welch_gpu[1024-1000000.0-16384-complex128] (0005_branch-)     580.1170 (1.08)   
test_welch_gpu[1024-1.0-16384-complex128] (0004_updates)           580.8460 (1.08)   
test_welch_gpu[2048-1.0-16384-complex128] (0005_branch-)           582.8700 (1.09)   
test_welch_gpu[1024-1.0-16384-complex128] (0005_branch-)           583.2727 (1.09)   
-------------------------------------------------------------------------------------


------------------ benchmark 'Wiener': 4 tests -------------------
Name (time in us)                                   Mean          
------------------------------------------------------------------
test_wiener_gpu[1-32768] (0004_updates)         366.4667 (1.0)    
test_wiener_gpu[32768] (0005_branch-)           493.6949 (1.35)   
test_wiener_gpu[2-256] (0004_updates)         1,276.0011 (3.48)   
test_wiener_gpu[16777216] (0005_branch-)     67,322.7580 (183.71) 
------------------------------------------------------------------

Legend:
  Outliers: 1 Standard Deviation from Mean; 1.5 IQR (InterQuartile Range) from 1st Quartile and 3rd Quartile.
  OPS: Operations Per Second, computed as 1 / Mean

@mnicely mnicely merged commit 30f2fa9 into rapidsai:branch-0.17 Oct 16, 2020
@mnicely mnicely deleted the add_c++11 branch October 16, 2020 15:51
This was referenced Oct 16, 2020
Closed
Closed
Closed
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Reviewers

@awthomp awthomp awthomp approved these changes

Assignees

@mnicely mnicely

Labels
3 - Ready for Review Ready for review by team
Projects
None yet
Milestone
0.17
Development

Successfully merging this pull request may close these issues.
4 participants
@mnicely @GPUtester @awthomp @BradReesWork