API tools faq
paste
Advertisement
Guest User

Untitled

a guest
Feb 8th, 2017
114
0
Never
Add comment
Not a member of Pastebin yet? Sign Up, it unlocks many cool features!
text 6.72 KB | None | 0 0
raw download clone embed print report
  1. (sklearn-env) [ec2-user@ip-172-31-11-89 ~]$ ipython pairwise_distances_benchmark.py True euclidean
  2. ================================================================================
  3.  
  4. # sparse=True, n_x=100000, n_y=1000, n_dim=10000
  5. # X array: 0.16 GB, Y array 0.0016 GB, result array 0.8 GB
  6. # metric = euclidean
  7. ================================================================================
  8.  
  9. ## {'MKL_NUM_THREADS': 8, 'mmap_result': False, 'backend': 'multiprocessing'}
  10. n_jobs= 1 => 1 loop, best of 1: 3.46 s per loop
  11. n_jobs= 2 => 1 loop, best of 1: 5.84 s per loop
  12. n_jobs= 4 => 1 loop, best of 1: 4.91 s per loop
  13. n_jobs= 8 => 1 loop, best of 1: 4.76 s per loop
  14. n_jobs= 16 => 1 loop, best of 1: 4.92 s per loop
  15.  
  16. ## {'MKL_NUM_THREADS': 8, 'mmap_result': True, 'backend': 'multiprocessing'}
  17. n_jobs= 1 => 1 loop, best of 1: 3.48 s per loop
  18. n_jobs= 2 => 1 loop, best of 1: 5.83 s per loop
  19. n_jobs= 4 => 1 loop, best of 1: 7.55 s per loop
  20. n_jobs= 8 => 1 loop, best of 1: 13 s per loop
  21. n_jobs= 16 => 1 loop, best of 1: 24 s per loop
  22.  
  23. ## {'MKL_NUM_THREADS': 8, 'backend': 'threading'}
  24. n_jobs= 1 => 1 loop, best of 1: 3.48 s per loop
  25. n_jobs= 2 => 1 loop, best of 1: 2.92 s per loop
  26. n_jobs= 4 => 1 loop, best of 1: 2.27 s per loop
  27. n_jobs= 8 => 1 loop, best of 1: 2.41 s per loop
  28. n_jobs= 16 => 1 loop, best of 1: 3.09 s per loop
  29.  
  30. ## {'MKL_NUM_THREADS': 1, 'mmap_result': False, 'backend': 'multiprocessing'}
  31. n_jobs= 1 => 1 loop, best of 1: 3.45 s per loop
  32. n_jobs= 2 => 1 loop, best of 1: 6.64 s per loop
  33. n_jobs= 4 => 1 loop, best of 1: 5.47 s per loop
  34. n_jobs= 8 => 1 loop, best of 1: 4.81 s per loop
  35. n_jobs= 16 => 1 loop, best of 1: 5.26 s per loop
  36.  
  37. ## {'MKL_NUM_THREADS': 1, 'mmap_result': True, 'backend': 'multiprocessing'}
  38. n_jobs= 1 => 1 loop, best of 1: 3.47 s per loop
  39. n_jobs= 2 => 1 loop, best of 1: 5.86 s per loop
  40. n_jobs= 4 => 1 loop, best of 1: 7.6 s per loop
  41. n_jobs= 8 => 1 loop, best of 1: 13 s per loop
  42. n_jobs= 16 => 1 loop, best of 1: 24.2 s per loop
  43.  
  44. ## {'MKL_NUM_THREADS': 1, 'backend': 'threading'}
  45. n_jobs= 1 => 1 loop, best of 1: 3.47 s per loop
  46. n_jobs= 2 => 1 loop, best of 1: 2.94 s per loop
  47. n_jobs= 4 => 1 loop, best of 1: 2.29 s per loop
  48. n_jobs= 8 => 1 loop, best of 1: 2.38 s per loop
  49. n_jobs= 16 => 1 loop, best of 1: 3.36 s per loop
  50. ================================================================================
  51.  
  52. # sparse=True, n_x=10000, n_y=10000, n_dim=10000
  53. # X array: 0.016 GB, Y array 0.016 GB, result array 0.8 GB
  54. # metric = euclidean
  55. ================================================================================
  56. ## {'MKL_NUM_THREADS': 8, 'mmap_result': False, 'backend': 'multiprocessing'}
  57. n_jobs= 1 => 1 loop, best of 1: 3.2 s per loop
  58. n_jobs= 2 => 1 loop, best of 1: 5.42 s per loop
  59. n_jobs= 4 => 1 loop, best of 1: 4.4 s per loop
  60. n_jobs= 8 => 1 loop, best of 1: 3.96 s per loop
  61. n_jobs= 16 => 1 loop, best of 1: 3.85 s per loop
  62.  
  63. ## {'MKL_NUM_THREADS': 8, 'mmap_result': True, 'backend': 'multiprocessing'}
  64. n_jobs= 1 => 1 loop, best of 1: 3.2 s per loop
  65. n_jobs= 2 => 1 loop, best of 1: 4.95 s per loop
  66. n_jobs= 4 => 1 loop, best of 1: 6.58 s per loop
  67. n_jobs= 8 => 1 loop, best of 1: 11.4 s per loop
  68. n_jobs= 16 => 1 loop, best of 1: 21.5 s per loop
  69.  
  70. ## {'MKL_NUM_THREADS': 8, 'backend': 'threading'}
  71. n_jobs= 1 => 1 loop, best of 1: 3.2 s per loop
  72. n_jobs= 2 => 1 loop, best of 1: 2.58 s per loop
  73. n_jobs= 4 => 1 loop, best of 1: 1.9 s per loop
  74. n_jobs= 8 => 1 loop, best of 1: 1.9 s per loop
  75. n_jobs= 16 => 1 loop, best of 1: 2.56 s per loop
  76.  
  77. ## {'MKL_NUM_THREADS': 1, 'mmap_result': False, 'backend': 'multiprocessing'}
  78. n_jobs= 1 => 1 loop, best of 1: 3.18 s per loop
  79. n_jobs= 2 => 1 loop, best of 1: 5.4 s per loop
  80. n_jobs= 4 => 1 loop, best of 1: 4.72 s per loop
  81. n_jobs= 8 => 1 loop, best of 1: 4.02 s per loop
  82. n_jobs= 16 => 1 loop, best of 1: 3.85 s per loop
  83.  
  84. ## {'MKL_NUM_THREADS': 1, 'mmap_result': True, 'backend': 'multiprocessing'}
  85. n_jobs= 1 => 1 loop, best of 1: 3.2 s per loop
  86. n_jobs= 2 => 1 loop, best of 1: 5.17 s per loop
  87. n_jobs= 4 => 1 loop, best of 1: 6.7 s per loop
  88. n_jobs= 8 => 1 loop, best of 1: 11.5 s per loop
  89. n_jobs= 16 => 1 loop, best of 1: 21.6 s per loop
  90.  
  91. ## {'MKL_NUM_THREADS': 1, 'backend': 'threading'}
  92. n_jobs= 1 => 1 loop, best of 1: 3.2 s per loop
  93. n_jobs= 2 => 1 loop, best of 1: 2.98 s per loop
  94. n_jobs= 4 => 1 loop, best of 1: 1.9 s per loop
  95. n_jobs= 8 => 1 loop, best of 1: 2 s per loop
  96. n_jobs= 16 => 1 loop, best of 1: 2.57 s per loop
  97. ================================================================================
  98.  
  99. # sparse=True, n_x=10000, n_y=10000, n_dim=100
  100. # X array: 0.00016 GB, Y array 0.00016 GB, result array 0.8 GB
  101. # metric = euclidean
  102. ================================================================================
  103.  
  104. ## {'MKL_NUM_THREADS': 8, 'mmap_result': False, 'backend': 'multiprocessing'}
  105. n_jobs= 1 => 1 loop, best of 1: 1.05 s per loop
  106. n_jobs= 2 => 1 loop, best of 1: 4.59 s per loop
  107. n_jobs= 4 => 1 loop, best of 1: 3.8 s per loop
  108. n_jobs= 8 => 1 loop, best of 1: 3.56 s per loop
  109. n_jobs= 16 => ^[[A^[[B1 loop, best of 1: 3.61 s per loop
  110.  
  111. ## {'MKL_NUM_THREADS': 8, 'mmap_result': True, 'backend': multiprocessing'}
  112. n_jobs= 1 => 1 loop, best of 1: 1.05 s per loop
  113. n_jobs= 2 => 1 loop, best of 1: 3.76 s per loop
  114. n_jobs= 4 => 1 loop, best of 1: 5.91 s per loop
  115. n_jobs= 8 => 1 loop, best of 1: 10.8 s per loop
  116. n_jobs= 16 => 1 loop, best of 1: 20.8 s per loop
  117.  
  118. ## {'MKL_NUM_THREADS': 8, 'backend': 'threading'}
  119. n_jobs= 1 => 1 loop, best of 1: 1.05 s per loop
  120. n_jobs= 2 => 1 loop, best of 1: 1.37 s per loop
  121. n_jobs= 4 => 1 loop, best of 1: 1.28 s per loop
  122. n_jobs= 8 => 1 loop, best of 1: 1.49 s per loop
  123. n_jobs= 16 => 1 loop, best of 1: 735 ms per loop
  124.  
  125. ## {'MKL_NUM_THREADS': 1, 'mmap_result': False, 'backend': 'multiprocessing'}
  126. n_jobs= 1 => 1 loop, best of 1: 1.05 s per loop
  127. n_jobs= 2 => 1 loop, best of 1: 4.26 s per loop
  128. n_jobs= 4 => 1 loop, best of 1: 3.78 s per loop
  129. n_jobs= 8 => 1 loop, best of 1: 3.57 s per loop
  130. n_jobs= 16 => 1 loop, best of 1: 3.59 s per loop
  131.  
  132. ## {'MKL_NUM_THREADS': 1, 'mmap_result': True, 'backend': 'multiprocessing'}
  133. n_jobs= 1 => 1 loop, best of 1: 1.06 s per loop
  134. n_jobs= 2 => 1 loop, best of 1: 3.76 s per loop
  135. n_jobs= 4 => 1 loop, best of 1: 6 s per loop
  136. n_jobs= 8 => 1 loop, best of 1: 10.9 s per loop
  137. n_jobs= 16 => 1 loop, best of 1: 20.8 s per loop
  138.  
  139. ## {'MKL_NUM_THREADS': 1, 'backend': 'threading'}
  140. n_jobs= 1 => 1 loop, best of 1: 1.05 s per loop
  141. n_jobs= 2 => 1 loop, best of 1: 1.37 s per loop
  142. n_jobs= 4 => 1 loop, best of 1: 1.29 s per loop
  143. n_jobs= 8 => 1 loop, best of 1: 1.49 s per loop
  144. n_jobs= 16 => 1 loop, best of 1: 736 ms per loop
Advertisement
Add Comment
Please, Sign In to add comment
Advertisement
Public Pastes
Advertisement
We use cookies for various purposes including analytics. By continuing to use Pastebin, you agree to our use of cookies as described in the Cookies Policy.  OK, I Understand
Not a member of Pastebin yet?
Sign Up, it unlocks many cool features!