Guest User

Untitled

a guest
Sep 18th, 2018
68
0
Never
Not a member of Pastebin yet? Sign Up, it unlocks many cool features!
text 11.09 KB | None | 0 0
  1. ──────────────────────────────────────────────────────────────────────────────────────
  2. Time Allocations
  3. ────────────────────── ───────────────────────
  4. Tot / % measured: 104s / 31.7% 10.8GiB / 5.58%
  5.  
  6. Section ncalls time %tot avg alloc %tot avg
  7. ──────────────────────────────────────────────────────────────────────────────────────
  8. *.[1] 10.7k 3.37s 10.2% 315μs 10.6MiB 1.70% 1.01KiB
  9. *. 10.6k 1.19s 3.63% 113μs 1.91MiB 0.31% -
  10. *[1] 12.7k 3.31s 10.1% 261μs 10.4MiB 1.67% -
  11. Knet.A_mul_Bt 12.5k 2.90s 8.83% 232μs 4.78MiB 0.77% -
  12. sum_outgrads 172k 2.21s 6.71% 12.8μs 17.1MiB 2.75% -
  13. * 12.9k 2.14s 6.51% 166μs 4.87MiB 0.79% -
  14. *[2] 12.5k 2.03s 6.17% 163μs 10.3MiB 1.66% -
  15. Knet.At_mul_B 12.5k 1.69s 5.13% 135μs 4.78MiB 0.77% -
  16. permutedims[1] 100 1.86s 5.67% 18.6ms 111MiB 17.9% 1.11MiB
  17. permutedims 100 1.78s 5.41% 17.8ms 111MiB 17.9% 1.11MiB
  18. *.[2] 10.6k 1.64s 5.00% 155μs 12.8MiB 2.07% 1.24KiB
  19. *. 10.6k 1.30s 3.95% 123μs 8.95MiB 1.44% -
  20. *. 10.9k 1.33s 4.05% 122μs 9.15MiB 1.47% -
  21. sum 3.70k 1.01s 3.09% 274μs 1.60MiB 0.26% -
  22. record 115k 960ms 2.92% 8.36μs 53.1MiB 8.57% -
  23. Knet.conv4 200 765ms 2.33% 3.82ms 603KiB 0.09% 3.02KiB
  24. +. 14.1k 736ms 2.24% 52.2μs 5.94MiB 0.96% -
  25. Knet.elu.[1] 2.80k 726ms 2.21% 259μs 1.71MiB 0.28% -
  26. Knet.eluback. 2.80k 611ms 1.86% 218μs 487KiB 0.08% -
  27. Knet.conv4[1] 200 711ms 2.16% 3.55ms 572KiB 0.09% 2.86KiB
  28. +.[2] 13.9k 663ms 2.02% 47.7μs 6.52MiB 1.05% -
  29. cat 11.4k 645ms 1.96% 56.8μs 7.22MiB 1.16% -
  30. sum[1] 3.60k 635ms 1.93% 176μs 4.17MiB 0.67% 1.19KiB
  31. Knet.dropout 4.00k 569ms 1.73% 142μs 1.10MiB 0.18% -
  32. Knet.elu. 2.80k 477ms 1.45% 170μs 487KiB 0.08% -
  33. Knet.rnnforw[2] 100 449ms 1.36% 4.49ms 2.46MiB 0.40% 25.2KiB
  34. Knet.dropout[1] 4.00k 434ms 1.32% 109μs 1.04MiB 0.17% -
  35. Knet.rnnforw 100 332ms 1.01% 3.32ms 2.50MiB 0.40% 25.6KiB
  36. cat[1] 11.3k 319ms 0.97% 28.3μs 10.4MiB 1.68% -
  37. reshape[1] 31.0k 283ms 0.86% 9.13μs 3.93MiB 0.63% -
  38. getindex 8.40k 274ms 0.83% 32.6μs 5.62MiB 0.91% -
  39. Knet.conv4[2] 100 235ms 0.71% 2.35ms 286KiB 0.05% 2.86KiB
  40. transpose[1] 7.00k 212ms 0.64% 30.3μs 1.60MiB 0.26% -
  41. transpose 7.20k 208ms 0.63% 28.9μs 1.43MiB 0.23% -
  42. reshape 31.7k 154ms 0.47% 4.85μs 3.18MiB 0.51% -
  43. cat[2] 4.80k 147ms 0.45% 30.6μs 5.83MiB 0.94% 1.24KiB
  44. Knet.cudnnSoftmaxForward[1] 3.50k 144ms 0.44% 41.2μs 4.11MiB 0.66% 1.20KiB
  45. Knet.cudnnSoftmaxForward 3.60k 119ms 0.36% 33.1μs 2.91MiB 0.47% -
  46. +.[1] 13.9k 114ms 0.35% 8.20μs 869KiB 0.14% -
  47. Knet.sigm.[1] 1.20k 73.4ms 0.22% 61.1μs 688KiB 0.11% -
  48. Knet.sigmback. 1.20k 29.6ms 0.09% 24.7μs 151KiB 0.02% -
  49. getindex[1] 8.10k 72.4ms 0.22% 8.94μs 972KiB 0.15% -
  50. cat[3] 2.10k 59.8ms 0.18% 28.5μs 3.83MiB 0.62% 1.87KiB
  51. cat[5] 1.70k 57.7ms 0.18% 33.9μs 4.42MiB 0.71% 2.66KiB
  52. cat[4] 1.90k 57.6ms 0.18% 30.3μs 4.13MiB 0.67% 2.22KiB
  53. cat[6] 1.50k 56.4ms 0.17% 37.6μs 4.42MiB 0.71% 3.01KiB
  54. -. 2.40k 55.0ms 0.17% 22.9μs 359KiB 0.06% -
  55. cat[7] 1.30k 52.2ms 0.16% 40.2μs 4.29MiB 0.69% 3.38KiB
  56. cat[8] 1.10k 47.4ms 0.14% 43.1μs 4.10MiB 0.66% 3.81KiB
  57. cat[62] 100 44.6ms 0.14% 446μs 4.61MiB 0.74% 47.2KiB
  58. cat[9] 900 41.2ms 0.13% 45.8μs 3.78MiB 0.61% 4.30KiB
  59. -.[2] 1.20k 38.6ms 0.12% 32.2μs 188KiB 0.03% -
  60. cat[10] 700 36.4ms 0.11% 52.0μs 3.32MiB 0.54% 4.86KiB
  61. cat[46] 100 35.2ms 0.11% 352μs 3.49MiB 0.56% 35.7KiB
  62. cat[54] 100 32.9ms 0.10% 329μs 4.05MiB 0.65% 41.4KiB
  63. cat[63] 100 30.5ms 0.09% 305μs 4.68MiB 0.75% 47.9KiB
  64. cat[59] 100 30.4ms 0.09% 304μs 4.40MiB 0.71% 45.0KiB
  65. cat[64] 100 30.1ms 0.09% 301μs 4.75MiB 0.77% 48.6KiB
  66. cat[61] 100 28.9ms 0.09% 289μs 4.54MiB 0.73% 46.5KiB
  67. cat[60] 100 28.9ms 0.09% 289μs 4.47MiB 0.72% 45.8KiB
  68. cat[58] 100 28.2ms 0.09% 282μs 4.33MiB 0.70% 44.3KiB
  69. cat[57] 100 27.7ms 0.08% 277μs 4.26MiB 0.69% 43.6KiB
  70. cat[11] 500 27.3ms 0.08% 54.6μs 2.73MiB 0.44% 5.58KiB
  71. cat[55] 100 27.2ms 0.08% 272μs 4.12MiB 0.66% 42.2KiB
  72. Knet.sigm. 1.20k 27.1ms 0.08% 22.6μs 151KiB 0.02% -
  73. cat[30] 100 27.1ms 0.08% 271μs 2.36MiB 0.38% 24.2KiB
  74. cat[56] 100 26.9ms 0.08% 269μs 4.19MiB 0.68% 42.9KiB
  75. cat[52] 100 25.7ms 0.08% 257μs 3.91MiB 0.63% 40.0KiB
  76. cat[53] 100 25.7ms 0.08% 257μs 3.98MiB 0.64% 40.7KiB
  77. cat[49] 100 25.2ms 0.08% 252μs 3.70MiB 0.60% 37.8KiB
  78. cat[51] 100 24.9ms 0.08% 249μs 3.84MiB 0.62% 39.3KiB
  79. cat[50] 100 24.4ms 0.07% 244μs 3.77MiB 0.61% 38.6KiB
  80. cat[47] 100 24.2ms 0.07% 242μs 3.56MiB 0.57% 36.4KiB
  81. cat[48] 100 24.1ms 0.07% 241μs 3.63MiB 0.58% 37.1KiB
  82. cat[43] 100 23.2ms 0.07% 232μs 3.27MiB 0.53% 33.5KiB
  83. cat[38] 100 22.9ms 0.07% 229μs 2.92MiB 0.47% 29.9KiB
  84. cat[45] 100 22.9ms 0.07% 229μs 3.42MiB 0.55% 35.0KiB
  85. cat[44] 100 22.7ms 0.07% 227μs 3.34MiB 0.54% 34.3KiB
  86. cat[42] 100 22.2ms 0.07% 222μs 3.20MiB 0.52% 32.8KiB
  87. cat[41] 100 21.9ms 0.07% 219μs 3.13MiB 0.51% 32.1KiB
  88. cat[40] 100 21.3ms 0.06% 213μs 3.06MiB 0.49% 31.4KiB
  89. cat[29] 100 20.3ms 0.06% 203μs 2.29MiB 0.37% 23.5KiB
  90. cat[37] 100 20.1ms 0.06% 201μs 2.85MiB 0.46% 29.2KiB
  91. cat[39] 100 19.8ms 0.06% 198μs 2.99MiB 0.48% 30.7KiB
  92. cat[35] 100 19.2ms 0.06% 192μs 2.71MiB 0.44% 27.8KiB
  93. cat[36] 100 18.9ms 0.06% 189μs 2.78MiB 0.45% 28.5KiB
  94. cat[12] 300 18.7ms 0.06% 62.2μs 2.02MiB 0.33% 6.91KiB
  95. cat[34] 100 18.6ms 0.06% 186μs 2.64MiB 0.43% 27.1KiB
  96. cat[33] 100 18.4ms 0.06% 184μs 2.57MiB 0.41% 26.3KiB
  97. cat[32] 100 18.3ms 0.06% 183μs 2.50MiB 0.40% 25.6KiB
  98. cat[31] 100 17.8ms 0.05% 178μs 2.43MiB 0.39% 24.9KiB
  99. cat[22] 100 17.4ms 0.05% 174μs 1.80MiB 0.29% 18.4KiB
  100. cat[28] 100 16.4ms 0.05% 164μs 2.22MiB 0.36% 22.8KiB
  101. cat[27] 100 15.8ms 0.05% 158μs 2.15MiB 0.35% 22.0KiB
  102. cat[14] 100 15.7ms 0.05% 157μs 1.24MiB 0.20% 12.7KiB
  103. cat[26] 100 15.6ms 0.05% 156μs 2.08MiB 0.34% 21.3KiB
  104. cat[25] 100 15.0ms 0.05% 150μs 2.01MiB 0.32% 20.6KiB
  105. cat[15] 100 15.0ms 0.05% 150μs 1.31MiB 0.21% 13.4KiB
  106. cat[24] 100 14.8ms 0.04% 148μs 1.94MiB 0.31% 19.9KiB
  107. cat[23] 100 14.5ms 0.04% 145μs 1.87MiB 0.30% 19.2KiB
  108. cat[21] 100 13.4ms 0.04% 134μs 1.73MiB 0.28% 17.7KiB
  109. cat[20] 100 13.0ms 0.04% 130μs 1.66MiB 0.27% 17.0KiB
  110. cat[19] 100 12.7ms 0.04% 127μs 1.59MiB 0.26% 16.3KiB
  111. cat[18] 100 12.3ms 0.04% 123μs 1.52MiB 0.25% 15.6KiB
  112. cat[17] 100 12.2ms 0.04% 122μs 1.45MiB 0.23% 14.8KiB
  113. Knet._logp[1] 100 12.0ms 0.04% 120μs 134KiB 0.02% 1.34KiB
  114. exp. 100 2.58ms 0.01% 25.8μs 17.3KiB 0.00% -
  115. cat[16] 100 11.5ms 0.04% 115μs 1.38MiB 0.22% 14.1KiB
  116. -.[1] 1.10k 11.3ms 0.03% 10.3μs 68.8KiB 0.01% -
  117. cat[13] 100 10.3ms 0.03% 103μs 1.17MiB 0.19% 12.0KiB
  118. Knet._logp 100 9.62ms 0.03% 96.2μs 159KiB 0.02% 1.59KiB
  119. -[1] 100 2.18ms 0.01% 21.8μs 9.38KiB 0.00% -
  120. Knet.rnnforw[3] 100 1.59ms 0.00% 15.9μs 7.81KiB 0.00% -
  121. - 100 461μs 0.00% 4.61μs 6.25KiB 0.00% -
  122. ──────────────────────────────────────────────────────────────────────────────────────
Add Comment
Please, Sign In to add comment