Skip to content

BENCH: add correlate/convolve benchmarks. #7689

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom
Merged
merged 1 commit into from
Jun 3, 2016
Merged

BENCH: add correlate/convolve benchmarks. #7689

merged 1 commit into from
Jun 3, 2016

Conversation

rgommers
Copy link
Member

Related to gh-5978, which makes significant changes to these functions.

Output:

[  0.00%] ·· Benchmarking /usr/bin/python
[ 50.00%] ··· Running bench_core.CorrConv.time_convolve                                              ok
[ 50.00%] ···· 
               ========== ========= ========== ========== ==========
               --                                 mode              
               -------------------- --------------------------------
                 size1      size2     valid       same       full   
               ========== ========= ========== ========== ==========
                   50         10     46.30μs    31.43μs    32.99μs  
                   50        100     48.34μs    56.88μs    78.45μs  
                   50        1000    464.67μs   259.96μs   510.50μs 
                   50      10000.0    4.06ms     3.82ms     3.03ms  
                  1000        10     123.68μs   21.80μs    130.21μs 
                  1000       100     181.35μs   629.60μs   470.75μs 
                  1000       1000    48.17μs     2.80ms     3.79ms  
                  1000     10000.0   28.42ms    11.40ms    32.60ms  
                100000.0      10      6.96ms     2.68ms     9.54ms  
                100000.0     100     24.32ms    35.93ms    27.21ms  
                100000.0     1000    20.84ms    21.34ms    21.35ms  
                100000.0   10000.0   225.43ms   243.65ms   253.61ms 
               ========== ========= ========== ========== ==========

[100.00%] ··· Running bench_core.CorrConv.time_correlate                                             ok
[100.00%] ···· 
               ========== ========= ========== ========== ==========
               --                                 mode              
               -------------------- --------------------------------
                 size1      size2     valid       same       full   
               ========== ========= ========== ========== ==========
                   50         10     30.45μs    38.19μs    19.76μs  
                   50        100     90.23μs    97.56μs    115.63μs 
                   50        1000    256.14μs   206.78μs   519.70μs 
                   50      10000.0    3.11ms     2.65ms     2.13ms  
                  1000        10     38.81μs    87.11μs    69.46μs  
                  1000       100     277.78μs   615.10μs   310.64μs 
                  1000       1000    42.39μs     2.58ms     3.77ms  
                  1000     10000.0   29.27ms    11.78ms    31.05ms  
                100000.0      10      9.39ms     9.35ms     8.62ms  
                100000.0     100     29.83ms    52.04ms    51.65ms  
                100000.0     1000    146.30ms   21.51ms    20.61ms  
                100000.0   10000.0   232.52ms   246.62ms   253.96ms 
               ========== ========= ========== ========== ==========

@rgommers
BENCH: add correlate/convolve benchmarks.
af13636 
Related to numpygh-5978, which makes significant changes to these functions.
@rgommers rgommers mentioned this pull request May 28, 2016
Closed
@charris charris merged commit a55b2dd into numpy:master Jun 3, 2016
@charris
Copy link
Member

charris commented Jun 3, 2016

Thanks Ralf.

@homu homu mentioned this pull request Jun 3, 2016
Merged
@rgommers rgommers deleted the correlate-benchmarks branch June 4, 2016 12:33
@HoniSanders
Copy link

How do I run these benchmarks?

@rgommers
Copy link
Member Author

The easiest is python runtests.py --bench bench_core. See benchmarks/README for more details.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers
No reviews
Assignees
No one assigned
Labels
component: benchmarks component: numpy._core
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
3 participants
@rgommers @charris @HoniSanders