API tools faq
paste
Advertisement
Guest User

Untitled

a guest
Feb 8th, 2017
204
0
Never
Add comment
Not a member of Pastebin yet? Sign Up, it unlocks many cool features!
text 6.16 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 False euclidean
  2.  
  3. # sparse=False, n_x=100000, n_y=1000, n_dim=1000
  4. # X array: 0.8 GB, Y array 0.008 GB, result array 0.8 GB
  5. # metric = euclidean
  6.  
  7. ## {'backend': 'multiprocessing', 'mmap_result': False, 'MKL_NUM_THREADS': 8}
  8. n_jobs= 1 => 1 loop, best of 3: 2.32 s per loop
  9. n_jobs= 2 => 1 loop, best of 3: 6.76 s per loop
  10. n_jobs= 4 => 1 loop, best of 3: 7.74 s per loop
  11. n_jobs= 8 => 1 loop, best of 3: 12.2 s per loop
  12. n_jobs= 16 => 1 loop, best of 3: 21.9 s per loop
  13.  
  14. ## {'backend': 'multiprocessing', 'mmap_result': True, 'MKL_NUM_THREADS': 8}
  15. n_jobs= 1 => 1 loop, best of 3: 1.76 s per loop
  16. n_jobs= 2 => 1 loop, best of 3: 7.52 s per loop
  17. n_jobs= 4 => 1 loop, best of 3: 12 s per loop
  18. n_jobs= 8 => 1 loop, best of 3: 21.6 s per loop
  19. n_jobs= 16 => 1 loop, best of 3: 41.4 s per loop
  20.  
  21. ## {'backend': 'threading', 'MKL_NUM_THREADS': 8}
  22. n_jobs= 1 => 1 loop, best of 3: 1.76 s per loop
  23. n_jobs= 2 => 1 loop, best of 3: 1.97 s per loop
  24. n_jobs= 4 => 1 loop, best of 3: 1.82 s per loop
  25. n_jobs= 8 => 1 loop, best of 3: 2.1 s per loop
  26. n_jobs= 16 => 1 loop, best of 3: 2.69 s per loop
  27.  
  28. ## {'backend': 'multiprocessing', 'mmap_result': False, 'MKL_NUM_THREADS': 1}
  29. n_jobs= 1 => 1 loop, best of 3: 1.77 s per loop
  30. n_jobs= 2 => 1 loop, best of 3: 6.78 s per loop
  31. n_jobs= 4 => 1 loop, best of 3: 7.82 s per loop
  32. n_jobs= 8 => 1 loop, best of 3: 12.2 s per loop
  33. n_jobs= 16 => 1 loop, best of 3: 22 s per loop
  34.  
  35. ## {'backend': 'multiprocessing', 'mmap_result': True, 'MKL_NUM_THREADS': 1}
  36. n_jobs= 1 => 1 loop, best of 3: 2.34 s per loop
  37. n_jobs= 2 => 1 loop, best of 3: 7.23 s per loop
  38. n_jobs= 4 => 1 loop, best of 3: 11.8 s per loop
  39. n_jobs= 8 => 1 loop, best of 3: 21.6 s per loop
  40. n_jobs= 16 => 1 loop, best of 3: 41.4 s per loop
  41.  
  42. ## {'backend': 'threading', 'MKL_NUM_THREADS': 1}
  43. n_jobs= 1 => 1 loop, best of 3: 2.34 s per loop
  44. n_jobs= 2 => 1 loop, best of 3: 1.92 s per loop
  45. n_jobs= 4 => 1 loop, best of 3: 1.83 s per loop
  46. n_jobs= 8 => 1 loop, best of 3: 2.12 s per loop
  47. n_jobs= 16 => 1 loop, best of 3: 2.68 s per loop
  48.  
  49. # sparse=False, n_x=10000, n_y=10000, n_dim=1000
  50. # X array: 0.08 GB, Y array 0.08 GB, result array 0.8 GB
  51. # metric = euclidean
  52.  
  53. ## {'backend': 'multiprocessing', 'mmap_result': False, 'MKL_NUM_THREADS': 8}
  54. n_jobs= 1 => 1 loop, best of 3: 2.17 s per loop
  55. n_jobs= 2 => 1 loop, best of 3: 5.33 s per loop
  56. n_jobs= 4 => 1 loop, best of 3: 4.77 s per loop
  57. n_jobs= 8 => 1 loop, best of 3: 4.35 s per loop
  58. n_jobs= 16 => 1 loop, best of 3: 4.81 s per loop
  59.  
  60. ## {'backend': 'multiprocessing', 'mmap_result': True, 'MKL_NUM_THREADS': 8}
  61. n_jobs= 1 => 1 loop, best of 3: 1.6 s per loop
  62. n_jobs= 2 => 1 loop, best of 3: 4.74 s per loop
  63. n_jobs= 4 => 1 loop, best of 3: 7.26 s per loop
  64. n_jobs= 8 => 1 loop, best of 3: 12.9 s per loop
  65. n_jobs= 16 => 1 loop, best of 3: 24.8 s per loop
  66.  
  67. ## {'backend': 'threading', 'MKL_NUM_THREADS': 8}
  68. n_jobs= 1 => 1 loop, best of 3: 2.21 s per loop
  69. n_jobs= 2 => 1 loop, best of 3: 1.77 s per loop
  70. n_jobs= 4 => 1 loop, best of 3: 1.62 s per loop
  71. n_jobs= 8 => 1 loop, best of 3: 1.62 s per loop
  72. n_jobs= 16 => 1 loop, best of 3: 1.44 s per loop
  73.  
  74. ## {'backend': 'multiprocessing', 'mmap_result': False, 'MKL_NUM_THREADS': 1}
  75. n_jobs= 1 => 1 loop, best of 3: 1.6 s per loop
  76. n_jobs= 2 => 1 loop, best of 3: 5.41 s per loop
  77. n_jobs= 4 => 1 loop, best of 3: 4.6 s per loop
  78. n_jobs= 8 => 1 loop, best of 3: 4.55 s per loop
  79. n_jobs= 16 => 1 loop, best of 3: 4.88 s per loop
  80.  
  81. ## {'backend': 'multiprocessing', 'mmap_result': True, 'MKL_NUM_THREADS': 1}
  82. n_jobs= 1 => 1 loop, best of 3: 2.21 s per loop
  83. n_jobs= 2 => 1 loop, best of 3: 4.74 s per loop
  84. n_jobs= 4 => 1 loop, best of 3: 7.27 s per loop
  85. n_jobs= 8 => 1 loop, best of 3: 12.9 s per loop
  86. n_jobs= 16 => 1 loop, best of 3: 24.8 s per loop
  87.  
  88. ## {'backend': 'threading', 'MKL_NUM_THREADS': 1}
  89. n_jobs= 1 => 1 loop, best of 3: 1.6 s per loop
  90. n_jobs= 2 => 1 loop, best of 3: 1.77 s per loop
  91. n_jobs= 4 => 1 loop, best of 3: 1.62 s per loop
  92. n_jobs= 8 => 1 loop, best of 3: 1.57 s per loop
  93. n_jobs= 16 => 1 loop, best of 3: 1.42 s per loop
  94.  
  95. # sparse=False, n_x=10000, n_y=10000, n_dim=10
  96. # X array: 0.0008 GB, Y array 0.0008 GB, result array 0.8 GB
  97. # metric = euclidean
  98.  
  99. ## {'backend': 'multiprocessing', 'mmap_result': False, 'MKL_NUM_THREADS': 8}
  100. n_jobs= 1 => 1 loop, best of 3: 1 s per loop
  101. n_jobs= 2 => 1 loop, best of 3: 4.48 s per loop
  102. n_jobs= 4 => 1 loop, best of 3: 3.89 s per loop
  103. n_jobs= 8 => 1 loop, best of 3: 3.78 s per loop
  104. n_jobs= 16 => 1 loop, best of 3: 3.86 s per loop
  105.  
  106. ## {'backend': 'multiprocessing', 'mmap_result': True, 'MKL_NUM_THREADS': 8}
  107. n_jobs= 1 => 1 loop, best of 3: 995 ms per loop
  108. n_jobs= 2 => 1 loop, best of 3: 3.72 s per loop
  109. n_jobs= 4 => 1 loop, best of 3: 5.95 s per loop
  110. n_jobs= 8 => 1 loop, best of 3: 10.7 s per loop
  111. n_jobs= 16 => 1 loop, best of 3: 20.6 s per loop
  112.  
  113. ## {'backend': 'threading', 'MKL_NUM_THREADS': 8}
  114. n_jobs= 1 => 1 loop, best of 3: 1.01 s per loop
  115. n_jobs= 2 => 1 loop, best of 3: 1.13 s per loop
  116. n_jobs= 4 => 1 loop, best of 3: 1.07 s per loop
  117. n_jobs= 8 => 1 loop, best of 3: 996 ms per loop
  118. n_jobs= 16 => 1 loop, best of 3: 1.03 s per loop
  119.  
  120. ## {'backend': 'multiprocessing', 'mmap_result': False, 'MKL_NUM_THREADS': 1}
  121. n_jobs= 1 => 1 loop, best of 3: 998 ms per loop
  122. n_jobs= 2 => 1 loop, best of 3: 4.31 s per loop
  123. n_jobs= 4 => 1 loop, best of 3: 3.93 s per loop
  124. n_jobs= 8 => 1 loop, best of 3: 3.79 s per loop
  125. n_jobs= 16 => 1 loop, best of 3: 3.88 s per loop
  126.  
  127. ## {'backend': 'multiprocessing', 'mmap_result': True, 'MKL_NUM_THREADS': 1}
  128. n_jobs= 1 => 1 loop, best of 3: 1 s per loop
  129. n_jobs= 2 => 1 loop, best of 3: 3.72 s per loop
  130. n_jobs= 4 => 1 loop, best of 3: 5.85 s per loop
  131. n_jobs= 8 => 1 loop, best of 3: 10.7 s per loop
  132. n_jobs= 16 => 1 loop, best of 3: 20.6 s per loop
  133.  
  134. ## {'backend': 'threading', 'MKL_NUM_THREADS': 1}
  135. n_jobs= 1 => 1 loop, best of 3: 1.04 s per loop
  136. n_jobs= 2 => 1 loop, best of 3: 1.13 s per loop
  137. n_jobs= 4 => 1 loop, best of 3: 1.08 s per loop
  138. n_jobs= 8 => 1 loop, best of 3: 991 ms per loop
  139. n_jobs= 16 => 1 loop, best of 3: 1.02 s 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!