Dandandan opened a new pull request #9759:
URL: https://github.com/apache/arrow/pull/9759


   This adds a function `from_trusted_len_iter_bool` to speed up the creation 
of an array for booleans.
   
   Benchmarks are a bit noisy, but seems to be ~10-20% faster for comparison 
kernels.
   
   ```
   Gnuplot not found, using plotters backend
   eq Float32              time:   [54.204 us 54.284 us 54.364 us]              
         
                           change: [-29.087% -28.838% -28.581%] (p = 0.00 < 
0.05)
                           Performance has improved.
   Found 6 outliers among 100 measurements (6.00%)
     5 (5.00%) low mild
     1 (1.00%) high mild
   
   eq scalar Float32       time:   [43.660 us 43.743 us 43.830 us]              
                 
                           change: [-30.819% -30.545% -30.269%] (p = 0.00 < 
0.05)
                           Performance has improved.
   Found 5 outliers among 100 measurements (5.00%)
     4 (4.00%) high mild
     1 (1.00%) high severe
   
   neq Float32             time:   [68.726 us 68.893 us 69.048 us]              
          
                           change: [-14.045% -13.772% -13.490%] (p = 0.00 < 
0.05)
                           Performance has improved.
   Found 1 outliers among 100 measurements (1.00%)
     1 (1.00%) high mild
   
   neq scalar Float32      time:   [46.251 us 46.322 us 46.395 us]              
                  
                           change: [-12.204% -11.952% -11.702%] (p = 0.00 < 
0.05)
                           Performance has improved.
   Found 6 outliers among 100 measurements (6.00%)
     1 (1.00%) low mild
     5 (5.00%) high mild
   
   lt Float32              time:   [50.264 us 50.438 us 50.613 us]              
          
                           change: [-21.300% -20.964% -20.649%] (p = 0.00 < 
0.05)
                           Performance has improved.
   
   lt scalar Float32       time:   [48.847 us 48.929 us 49.013 us]              
                 
                           change: [-10.132% -9.9180% -9.6910%] (p = 0.00 < 
0.05)
                           Performance has improved.
   Found 5 outliers among 100 measurements (5.00%)
     4 (4.00%) high mild
     1 (1.00%) high severe
   
   lt_eq Float32           time:   [46.105 us 46.198 us 46.282 us]              
             
                           change: [-21.276% -20.966% -20.703%] (p = 0.00 < 
0.05)
                           Performance has improved.
   Found 18 outliers among 100 measurements (18.00%)
     2 (2.00%) low severe
     13 (13.00%) low mild
     1 (1.00%) high mild
     2 (2.00%) high severe
   
   lt_eq scalar Float32    time:   [47.359 us 47.456 us 47.593 us]              
                    
                           change: [+0.2766% +0.5240% +0.7821%] (p = 0.00 < 
0.05)
                           Change within noise threshold.
   Found 10 outliers among 100 measurements (10.00%)
     8 (8.00%) high mild
     2 (2.00%) high severe
   
   gt Float32              time:   [57.313 us 57.363 us 57.412 us]              
         
                           change: [-18.328% -18.177% -18.031%] (p = 0.00 < 
0.05)
                           Performance has improved.
   Found 3 outliers among 100 measurements (3.00%)
     2 (2.00%) low severe
     1 (1.00%) low mild
   
   gt scalar Float32       time:   [44.091 us 44.132 us 44.175 us]              
                 
                           change: [-9.4233% -9.2747% -9.1273%] (p = 0.00 < 
0.05)
                           Performance has improved.
   Found 7 outliers among 100 measurements (7.00%)
     4 (4.00%) low mild
     3 (3.00%) high mild
   
   gt_eq Float32           time:   [55.856 us 55.932 us 56.007 us]              
            
                           change: [-7.4997% -7.2656% -7.0334%] (p = 0.00 < 
0.05)
                           Performance has improved.
   Found 3 outliers among 100 measurements (3.00%)
     1 (1.00%) low mild
     2 (2.00%) high mild
   
   gt_eq scalar Float32    time:   [42.365 us 42.419 us 42.482 us]              
                    
                           change: [+0.5289% +0.7174% +0.9116%] (p = 0.00 < 
0.05)
                           Change within noise threshold.
   Found 5 outliers among 100 measurements (5.00%)
     2 (2.00%) high mild
     3 (3.00%) high severe
   ```


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
[email protected]


Reply via email to