Advertisement
Guest User

Untitled

a guest
Mar 18th, 2019
67
0
Never
Not a member of Pastebin yet? Sign Up, it unlocks many cool features!
text 44.04 KB | None | 0 0
  1. WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/tensorflow/python/ops/math_ops.py:3066: to_int32 (from tensorflow.python.ops.math_ops) is deprecated and will be removed in a future version.
  2. Instructions for updating:
  3. Use tf.cast instead.
  4. Using 39 evaluation batches
  5. 2019-03-18 17:57:18.932490: I tensorflow/stream_executor/dso_loader.cc:152] successfully opened CUDA library libcublas.so.10.0 locally
  6. step 1, policy=7.53778 value=1.10619 policy accuracy=0.010016% value accuracy=29.8427% mse=0.179107
  7. step 1, policy=7.52524 value=1.06444 policy accuracy=0.010016% value accuracy=60.7071% mse=0.156899
  8. step 2, lr=0.02 policy=7.68254 value=1.66804 mse=0.135876 reg=0.163218 total=9.5138 (176.806 pos/s)
  9. step 2, policy=7.51665 value=0.980072 policy accuracy=0.0350561% value accuracy=64.8888% mse=0.137208
  10. step 3, policy=7.50986 value=0.985969 policy accuracy=0.0651042% value accuracy=71.1689% mse=0.139408
  11. step 4, lr=0.02 policy=7.56576 value=0.810579 mse=0.0590109 reg=0.163213 total=8.53955 (506.731 pos/s)
  12. step 4, policy=7.50078 value=0.950471 policy accuracy=0.0450721% value accuracy=69.972% mse=0.120033
  13. step 5, policy=7.48683 value=0.92277 policy accuracy=0.155248% value accuracy=70.4978% mse=0.114326
  14. step 6, lr=0.02 policy=7.42792 value=0.5321 mse=0.0405634 reg=0.163218 total=8.12324 (506.1 pos/s)
  15. step 6, policy=7.4694 value=0.898504 policy accuracy=0.0851362% value accuracy=70.7181% mse=0.10828
  16. step 7, policy=7.44857 value=0.876742 policy accuracy=0.115184% value accuracy=70.7232% mse=0.102873
  17. step 8, lr=0.02 policy=7.25337 value=0.47918 mse=0.0380972 reg=0.16323 total=7.89578 (515.218 pos/s)
  18. step 8, policy=7.42086 value=0.853883 policy accuracy=0.385617% value accuracy=70.8383% mse=0.0969429
  19. step 9, policy=7.38639 value=0.828129 policy accuracy=1.0617% value accuracy=71.0086% mse=0.0905123
  20. step 10, lr=0.02 policy=7.03373 value=0.436518 mse=0.0346108 reg=0.163248 total=7.6335 (507.174 pos/s)
  21. step 10, policy=7.35061 value=0.820444 policy accuracy=1.20192% value accuracy=70.1773% mse=0.0867931
  22. step 11, policy=7.31163 value=0.806551 policy accuracy=1.252% value accuracy=70.012% mse=0.0809929
  23. step 12, lr=0.02 policy=6.81539 value=0.371434 mse=0.0288454 reg=0.163273 total=7.3501 (500.441 pos/s)
  24. step 12, policy=7.26945 value=0.814686 policy accuracy=1.19692% value accuracy=70.4377% mse=0.0857565
  25. step 13, policy=7.22599 value=0.835945 policy accuracy=1.27204% value accuracy=69.8968% mse=0.0986917
  26. step 14, lr=0.02 policy=6.60949 value=0.349712 mse=0.0270809 reg=0.163302 total=7.12251 (508.578 pos/s)
  27. step 14, policy=7.18353 value=0.876568 policy accuracy=1.29708% value accuracy=61.268% mse=0.107744
  28. step 15, policy=7.12877 value=0.917549 policy accuracy=1.35216% value accuracy=56.6506% mse=0.11227
  29. step 16, lr=0.02 policy=6.39876 value=0.327016 mse=0.0254125 reg=0.163336 total=6.88911 (513.621 pos/s)
  30. step 16, policy=7.0741 value=1.07214 policy accuracy=1.17188% value accuracy=36.6486% mse=0.170333
  31. step 17, policy=7.02364 value=0.985786 policy accuracy=1.47236% value accuracy=38.9523% mse=0.136761
  32. step 18, lr=0.02 policy=6.2792 value=0.582543 mse=0.0404752 reg=0.163373 total=7.02511 (507.277 pos/s)
  33. step 18, policy=6.97637 value=1.46531 policy accuracy=0.791266% value accuracy=37.1695% mse=0.258572
  34. step 19, policy=6.93345 value=1.67253 policy accuracy=0.615986% value accuracy=37.0944% mse=0.306898
  35. step 20, lr=0.02 policy=6.0593 value=0.332334 mse=0.0249982 reg=0.16341 total=6.55504 (508.621 pos/s)
  36. step 20, policy=6.87814 value=1.64682 policy accuracy=0.761218% value accuracy=36.6937% mse=0.305644
  37. step 21, policy=6.83539 value=1.63109 policy accuracy=0.73117% value accuracy=36.7388% mse=0.30121
  38. step 22, lr=0.02 policy=5.89063 value=0.250034 mse=0.0192065 reg=0.16345 total=6.30411 (504.16 pos/s)
  39. step 22, policy=6.78395 value=1.57007 policy accuracy=0.676082% value accuracy=37.2746% mse=0.29395
  40. step 23, policy=6.74483 value=1.6514 policy accuracy=0.746194% value accuracy=36.859% mse=0.309996
  41. step 24, lr=0.02 policy=5.82926 value=0.29842 mse=0.0232937 reg=0.163491 total=6.29117 (502.612 pos/s)
  42. step 24, policy=6.68111 value=1.19176 policy accuracy=0.686098% value accuracy=36.8339% mse=0.202243
  43. step 25, policy=6.6361 value=1.1721 policy accuracy=0.711138% value accuracy=37.3648% mse=0.214794
  44. Model saved in file: ./networks/net-64x6/net-64x6-25
  45. saved as './networks/net-64x6/net-64x6-25' 6.06M
  46. Weights saved in file: ./networks/net-64x6/net-64x6-25
  47. step 26, lr=0.02 policy=5.65041 value=0.230325 mse=0.0174988 reg=0.163533 total=6.04427 (443.469 pos/s)
  48. step 26, policy=6.59549 value=1.09681 policy accuracy=0.926482% value accuracy=37.2646% mse=0.191712
  49. step 27, policy=6.5535 value=1.12249 policy accuracy=0.95653% value accuracy=36.9692% mse=0.204199
  50. step 28, lr=0.02 policy=5.50016 value=0.199608 mse=0.0148712 reg=0.163576 total=5.86335 (505.182 pos/s)
  51. step 28, policy=6.51168 value=1.02566 policy accuracy=1.26202% value accuracy=39.7286% mse=0.175814
  52. step 29, policy=6.46305 value=0.949127 policy accuracy=1.63261% value accuracy=45.1823% mse=0.154848
  53. step 30, lr=0.02 policy=5.3242 value=0.194213 mse=0.0144739 reg=0.16362 total=5.68203 (509.375 pos/s)
  54. step 30, policy=6.40661 value=0.787763 policy accuracy=1.94812% value accuracy=61.0677% mse=0.0964814
  55. step 31, policy=6.36652 value=0.771961 policy accuracy=2.47396% value accuracy=60.2414% mse=0.0981387
  56. step 32, lr=0.02 policy=5.14259 value=0.175073 mse=0.0131327 reg=0.163666 total=5.48133 (507.458 pos/s)
  57. step 32, policy=6.31948 value=0.777881 policy accuracy=2.60917% value accuracy=58.2582% mse=0.104411
  58. step 33, policy=6.28399 value=0.768383 policy accuracy=3.25521% value accuracy=58.6388% mse=0.101112
  59. step 34, lr=0.02 policy=5.01352 value=0.165872 mse=0.0122376 reg=0.163714 total=5.34311 (504.458 pos/s)
  60. step 34, policy=6.24002 value=0.848758 policy accuracy=3.76603% value accuracy=52.8345% mse=0.134857
  61. step 35, policy=6.15393 value=0.591138 policy accuracy=5.45873% value accuracy=71.2891% mse=0.0466026
  62. step 36, lr=0.02 policy=4.75125 value=0.154373 mse=0.0113304 reg=0.163765 total=5.06939 (500.428 pos/s)
  63. step 36, policy=6.0864 value=0.577098 policy accuracy=7.04627% value accuracy=71.4543% mse=0.0448177
  64. step 37, policy=6.04008 value=0.644941 policy accuracy=6.73578% value accuracy=71.0236% mse=0.0735184
  65. step 38, lr=0.02 policy=4.63517 value=0.155799 mse=0.0115155 reg=0.16382 total=4.95479 (512.57 pos/s)
  66. step 38, policy=5.90972 value=0.654255 policy accuracy=11.5585% value accuracy=72.2356% mse=0.0497501
  67. step 39, policy=5.79358 value=0.530295 policy accuracy=12.6803% value accuracy=74.374% mse=0.0385881
  68. step 40, lr=0.02 policy=4.34273 value=0.122708 mse=0.00901117 reg=0.163878 total=4.62932 (507.073 pos/s)
  69. step 40, policy=5.78389 value=0.578737 policy accuracy=12.6302% value accuracy=78.4555% mse=0.0409307
  70. step 41, policy=5.60967 value=0.485426 policy accuracy=14.4131% value accuracy=81.1148% mse=0.0393558
  71. step 42, lr=0.02 policy=4.15636 value=0.18632 mse=0.0139359 reg=0.163939 total=4.50662 (502.627 pos/s)
  72. step 42, policy=5.57528 value=0.843827 policy accuracy=15.7302% value accuracy=77.7694% mse=0.0483309
  73. step 43, policy=5.31611 value=0.392982 policy accuracy=18.6398% value accuracy=82.507% mse=0.0268064
  74. step 44, lr=0.02 policy=4.02008 value=0.233508 mse=0.0168326 reg=0.164004 total=4.41759 (509.317 pos/s)
  75. step 44, policy=5.3544 value=0.681714 policy accuracy=17.2726% value accuracy=63.1761% mse=0.0549683
  76. step 45, policy=4.99792 value=0.895341 policy accuracy=26.0567% value accuracy=71.0387% mse=0.0613383
  77. step 46, lr=0.02 policy=3.72235 value=0.230055 mse=0.0164655 reg=0.16407 total=4.11647 (507.559 pos/s)
  78. step 46, policy=5.17903 value=1.15115 policy accuracy=19.0304% value accuracy=47.9167% mse=0.095598
  79. step 47, policy=4.69317 value=0.521734 policy accuracy=28.4155% value accuracy=76.7278% mse=0.0412209
  80. step 48, lr=0.02 policy=3.58002 value=0.166187 mse=0.0123045 reg=0.164138 total=3.91035 (497.424 pos/s)
  81. step 48, policy=4.65072 value=0.480538 policy accuracy=28.4605% value accuracy=82.3468% mse=0.0350345
  82. step 49, policy=4.53992 value=0.442461 policy accuracy=29.4371% value accuracy=83.5136% mse=0.0322837
  83. step 50, lr=0.02 policy=3.31473 value=0.100566 mse=0.00702034 reg=0.164208 total=3.57951 (507.815 pos/s)
  84. step 50, policy=4.43685 value=0.41909 policy accuracy=30.7292% value accuracy=80.1332% mse=0.0315073
  85. Model saved in file: ./networks/net-64x6/net-64x6-50
  86. saved as './networks/net-64x6/net-64x6-50' 5.98M
  87. Weights saved in file: ./networks/net-64x6/net-64x6-50
  88. step 51, policy=4.29827 value=0.360288 policy accuracy=31.9161% value accuracy=84.5202% mse=0.0262622
  89. step 52, lr=0.02 policy=3.12348 value=0.0997114 mse=0.00681202 reg=0.164278 total=3.38746 (442.594 pos/s)
  90. step 52, policy=4.17914 value=0.400881 policy accuracy=33.1881% value accuracy=84.0795% mse=0.0296148
  91. step 53, policy=4.02703 value=0.350266 policy accuracy=33.9643% value accuracy=84.7155% mse=0.0259661
  92. step 54, lr=0.02 policy=3.06248 value=0.0843017 mse=0.00554622 reg=0.164347 total=3.31112 (511.767 pos/s)
  93. step 54, policy=3.88576 value=0.392746 policy accuracy=35.1312% value accuracy=84.6554% mse=0.0291046
  94. step 55, policy=3.80428 value=0.41398 policy accuracy=35.2865% value accuracy=84.395% mse=0.0300156
  95. step 56, lr=0.02 policy=2.81421 value=0.0804113 mse=0.00553784 reg=0.164415 total=3.05903 (511.912 pos/s)
  96. step 56, policy=3.69207 value=0.359981 policy accuracy=36.1979% value accuracy=85.637% mse=0.0263362
  97. step 57, policy=3.54251 value=0.280447 policy accuracy=37.1394% value accuracy=87.7003% mse=0.0204565
  98. step 58, lr=0.02 policy=2.7422 value=0.0748301 mse=0.00519445 reg=0.16448 total=2.98151 (502.416 pos/s)
  99. step 58, policy=3.41021 value=0.311838 policy accuracy=38.0359% value accuracy=86.899% mse=0.023135
  100. step 59, policy=3.30605 value=0.242625 policy accuracy=38.742% value accuracy=89.0074% mse=0.0180555
  101. step 60, lr=0.02 policy=2.63323 value=0.0777168 mse=0.00534054 reg=0.164542 total=2.87549 (505.506 pos/s)
  102. step 60, policy=3.19044 value=0.393989 policy accuracy=39.4081% value accuracy=81.5254% mse=0.0314845
  103. step 61, policy=3.11065 value=0.333853 policy accuracy=39.6034% value accuracy=86.9742% mse=0.0245324
  104. step 62, lr=0.02 policy=2.52789 value=0.0657646 mse=0.00448108 reg=0.1646 total=2.75825 (511.065 pos/s)
  105. step 62, policy=3.02956 value=0.266262 policy accuracy=39.969% value accuracy=88.0459% mse=0.0207006
  106. step 63, policy=2.96171 value=0.348288 policy accuracy=40.7903% value accuracy=83.6438% mse=0.0277818
  107. step 64, lr=0.02 policy=2.49799 value=0.0665984 mse=0.00467111 reg=0.164654 total=2.72924 (500.061 pos/s)
  108. step 64, policy=2.8843 value=0.522595 policy accuracy=40.4197% value accuracy=83.4085% mse=0.0340896
  109. step 65, policy=2.84616 value=0.420473 policy accuracy=42.0072% value accuracy=80.7943% mse=0.0334858
  110. step 66, lr=0.02 policy=2.39166 value=0.0699372 mse=0.00475307 reg=0.164703 total=2.6263 (515.493 pos/s)
  111. step 66, policy=2.81583 value=0.317046 policy accuracy=41.1809% value accuracy=85.2815% mse=0.0253702
  112. step 67, policy=2.74249 value=0.532747 policy accuracy=42.0272% value accuracy=76.6376% mse=0.0422727
  113. step 68, lr=0.02 policy=2.27809 value=0.0633231 mse=0.0044323 reg=0.164749 total=2.50616 (510.32 pos/s)
  114. step 68, policy=2.63237 value=0.355739 policy accuracy=42.1274% value accuracy=86.1679% mse=0.0258333
  115. step 69, policy=2.62222 value=0.420406 policy accuracy=43.0138% value accuracy=81.4553% mse=0.0333796
  116. step 70, lr=0.02 policy=2.19847 value=0.0645817 mse=0.00447879 reg=0.164792 total=2.42785 (506.472 pos/s)
  117. step 70, policy=2.6084 value=0.319795 policy accuracy=43.2542% value accuracy=85.9024% mse=0.0250559
  118. step 71, policy=2.40135 value=0.236533 policy accuracy=44.8217% value accuracy=90.605% mse=0.0172743
  119. step 72, lr=0.02 policy=2.20218 value=0.0661815 mse=0.00474104 reg=0.164831 total=2.4332 (512.736 pos/s)
  120. step 72, policy=2.45061 value=0.273385 policy accuracy=45.1322% value accuracy=88.4315% mse=0.0210654
  121. step 73, policy=2.33809 value=0.162606 policy accuracy=46.1088% value accuracy=93.0188% mse=0.0122859
  122. step 74, lr=0.02 policy=2.08729 value=0.0596116 mse=0.00412107 reg=0.164867 total=2.31177 (508.828 pos/s)
  123. step 74, policy=2.32317 value=0.177649 policy accuracy=46.2039% value accuracy=92.3928% mse=0.0136921
  124. step 75, policy=2.27215 value=0.161754 policy accuracy=46.9501% value accuracy=93.119% mse=0.0123461
  125. Model saved in file: ./networks/net-64x6/net-64x6-75
  126. saved as './networks/net-64x6/net-64x6-75' 5.95M
  127. Weights saved in file: ./networks/net-64x6/net-64x6-75
  128. step 76, lr=0.02 policy=2.04251 value=0.0519772 mse=0.00362689 reg=0.164901 total=2.25939 (443.584 pos/s)
  129. step 76, policy=2.27298 value=0.275215 policy accuracy=46.6847% value accuracy=87.9808% mse=0.0215655
  130. step 77, policy=2.25246 value=0.117904 policy accuracy=47.1404% value accuracy=94.982% mse=0.00887731
  131. step 78, lr=0.02 policy=1.99812 value=0.04646 mse=0.00312821 reg=0.164932 total=2.20951 (506.982 pos/s)
  132. step 78, policy=2.18533 value=0.121454 policy accuracy=47.2155% value accuracy=94.972% mse=0.00905469
  133. step 79, policy=2.1691 value=0.061841 policy accuracy=47.9818% value accuracy=97.8215% mse=0.00412851
  134. step 80, lr=0.02 policy=1.9261 value=0.0491885 mse=0.00342753 reg=0.164961 total=2.14025 (510.304 pos/s)
  135. step 80, policy=2.10098 value=0.11438 policy accuracy=47.7063% value accuracy=95.613% mse=0.00821099
  136. step 81, policy=2.07885 value=0.0632724 policy accuracy=48.2522% value accuracy=97.6663% mse=0.00435834
  137. step 82, lr=0.02 policy=1.9403 value=0.044426 mse=0.00304882 reg=0.164989 total=2.14972 (504.225 pos/s)
  138. step 82, policy=2.01249 value=0.0437543 policy accuracy=49.2788% value accuracy=98.5427% mse=0.00282339
  139. step 83, policy=1.98942 value=0.0666784 policy accuracy=49.9048% value accuracy=97.6112% mse=0.00464875
  140. step 84, lr=0.02 policy=1.84681 value=0.0411681 mse=0.00278768 reg=0.165014 total=2.05299 (509.975 pos/s)
  141. step 84, policy=1.98546 value=0.0596794 policy accuracy=49.7396% value accuracy=97.6863% mse=0.00418005
  142. step 85, policy=1.93071 value=0.052021 policy accuracy=50.5609% value accuracy=98.147% mse=0.00358693
  143. step 86, lr=0.02 policy=1.88931 value=0.0432347 mse=0.00288478 reg=0.165038 total=2.09759 (508.727 pos/s)
  144. step 86, policy=1.92788 value=0.044741 policy accuracy=49.9599% value accuracy=98.5327% mse=0.00295392
  145. step 87, policy=1.91252 value=0.0480253 policy accuracy=50.3506% value accuracy=98.2672% mse=0.00329922
  146. step 88, lr=0.02 policy=1.83551 value=0.0511839 mse=0.00352739 reg=0.165061 total=2.05175 (500.46 pos/s)
  147. step 88, policy=1.90307 value=0.0555787 policy accuracy=50.4307% value accuracy=97.9768% mse=0.00386868
  148. step 89, policy=1.88934 value=0.0411775 policy accuracy=50.5909% value accuracy=98.5627% mse=0.00269993
  149. step 90, lr=0.02 policy=1.81916 value=0.052269 mse=0.00357635 reg=0.165082 total=2.03651 (504.594 pos/s)
  150. step 90, policy=1.8635 value=0.0667773 policy accuracy=50.5659% value accuracy=97.5962% mse=0.00459551
  151. step 91, policy=1.87593 value=0.0479916 policy accuracy=50.3506% value accuracy=98.3023% mse=0.00315376
  152. step 92, lr=0.02 policy=1.74743 value=0.0464162 mse=0.00328852 reg=0.165102 total=1.95895 (511.55 pos/s)
  153. step 92, policy=1.8374 value=0.0703165 policy accuracy=51.6226% value accuracy=97.2556% mse=0.00506748
  154. step 93, policy=1.79898 value=0.165063 policy accuracy=51.7879% value accuracy=93.8802% mse=0.0117986
  155. step 94, lr=0.02 policy=1.72195 value=0.0508075 mse=0.00355643 reg=0.165121 total=1.93787 (505.621 pos/s)
  156. step 94, policy=1.83932 value=0.104564 policy accuracy=51.7328% value accuracy=95.9936% mse=0.00761731
  157. step 95, policy=1.7918 value=0.446649 policy accuracy=50.8564% value accuracy=88.2462% mse=0.0251697
  158. step 96, lr=0.02 policy=1.73958 value=0.0467542 mse=0.00332313 reg=0.165139 total=1.95147 (514.539 pos/s)
  159. step 96, policy=1.74486 value=0.052871 policy accuracy=52.8446% value accuracy=98.112% mse=0.00361775
  160. step 97, policy=1.7385 value=0.18021 policy accuracy=52.0282% value accuracy=93.1941% mse=0.0129148
  161. step 98, lr=0.02 policy=1.69062 value=0.0560369 mse=0.00378822 reg=0.165157 total=1.91181 (512.697 pos/s)
  162. step 98, policy=1.74748 value=0.342544 policy accuracy=51.6026% value accuracy=89.2628% mse=0.0215577
  163. step 99, policy=1.69984 value=0.0420217 policy accuracy=52.7043% value accuracy=98.4926% mse=0.00285237
  164. step 100, lr=0.02 policy=1.7397 value=0.049633 mse=0.00348082 reg=0.165173 total=1.95451 (505.941 pos/s)
  165. step 100, policy=1.67084 value=0.0743241 policy accuracy=52.7744% value accuracy=97.0703% mse=0.00545764
  166. Model saved in file: ./networks/net-64x6/net-64x6-100
  167. saved as './networks/net-64x6/net-64x6-100' 5.94M
  168. Weights saved in file: ./networks/net-64x6/net-64x6-100
  169. step 101, policy=1.68714 value=0.058832 policy accuracy=52.6593% value accuracy=97.8966% mse=0.00406078
  170. step 102, lr=0.02 policy=1.68488 value=0.0481896 mse=0.00325521 reg=0.165188 total=1.89826 (449.636 pos/s)
  171. step 102, policy=1.6804 value=0.091346 policy accuracy=53.1951% value accuracy=96.249% mse=0.00671115
  172. step 103, policy=1.68169 value=0.302613 policy accuracy=53.0048% value accuracy=90.6751% mse=0.0188806
  173. step 104, lr=0.02 policy=1.64849 value=0.0660673 mse=0.00457472 reg=0.165203 total=1.87976 (501.507 pos/s)
  174. step 104, policy=1.66571 value=0.0585091 policy accuracy=53.4806% value accuracy=97.7965% mse=0.00415441
  175. step 105, policy=1.63039 value=0.0723669 policy accuracy=53.8862% value accuracy=97.1404% mse=0.00528145
  176. step 106, lr=0.02 policy=1.53316 value=0.0450167 mse=0.00299635 reg=0.165218 total=1.7434 (504.57 pos/s)
  177. step 106, policy=1.637 value=0.0720118 policy accuracy=53.2702% value accuracy=97.2957% mse=0.00515211
  178. step 107, policy=1.60882 value=0.0838754 policy accuracy=54.6074% value accuracy=96.7548% mse=0.00609099
  179. step 108, lr=0.02 policy=1.57193 value=0.0449297 mse=0.00306717 reg=0.165232 total=1.7821 (510.436 pos/s)
  180. step 108, policy=1.60796 value=0.100712 policy accuracy=53.2552% value accuracy=96.3542% mse=0.00705885
  181. step 109, policy=1.56958 value=0.0380662 policy accuracy=55.2634% value accuracy=98.6228% mse=0.00258797
  182. step 110, lr=0.02 policy=1.54007 value=0.0378862 mse=0.00249954 reg=0.165246 total=1.7432 (514.364 pos/s)
  183. step 110, policy=1.57891 value=0.075566 policy accuracy=54.1567% value accuracy=97.1905% mse=0.00540243
  184. step 111, policy=1.57445 value=0.0433777 policy accuracy=55.2434% value accuracy=98.4826% mse=0.0029436
  185. step 112, lr=0.02 policy=1.53775 value=0.0485105 mse=0.00338938 reg=0.16526 total=1.75152 (507.658 pos/s)
  186. step 112, policy=1.58181 value=0.116897 policy accuracy=54.2568% value accuracy=95.4026% mse=0.00853109
  187. step 113, policy=1.54508 value=0.0570491 policy accuracy=55.4437% value accuracy=97.9617% mse=0.00401926
  188. step 114, lr=0.02 policy=1.51263 value=0.0400143 mse=0.00250799 reg=0.165273 total=1.71791 (514.982 pos/s)
  189. step 114, policy=1.56745 value=0.0580586 policy accuracy=54.7977% value accuracy=97.7815% mse=0.00414557
  190. step 115, policy=1.57093 value=0.0602043 policy accuracy=54.6374% value accuracy=97.6713% mse=0.00427255
  191. step 116, lr=0.02 policy=1.44665 value=0.0380383 mse=0.00262076 reg=0.165286 total=1.64997 (506.977 pos/s)
  192. step 116, policy=1.56102 value=0.0755039 policy accuracy=54.8778% value accuracy=97.0252% mse=0.00553836
  193. step 117, policy=1.58027 value=0.0321666 policy accuracy=55.2835% value accuracy=98.7881% mse=0.00220443
  194. step 118, lr=0.02 policy=1.52117 value=0.044122 mse=0.00311072 reg=0.165299 total=1.73059 (504.708 pos/s)
  195. step 118, policy=1.49415 value=0.0494874 policy accuracy=56.255% value accuracy=98.107% mse=0.00347932
  196. step 119, policy=1.5395 value=0.0329377 policy accuracy=55.3786% value accuracy=98.8482% mse=0.0021909
  197. step 120, lr=0.02 policy=1.4788 value=0.0370093 mse=0.00251247 reg=0.165311 total=1.68112 (509.508 pos/s)
  198. step 120, policy=1.49101 value=0.0801662 policy accuracy=55.6791% value accuracy=97.1004% mse=0.00568101
  199. step 121, policy=1.49853 value=0.0300777 policy accuracy=56.0597% value accuracy=98.9884% mse=0.00203577
  200. step 122, lr=0.02 policy=1.46171 value=0.0383133 mse=0.00263127 reg=0.165323 total=1.66534 (518.177 pos/s)
  201. step 122, policy=1.4778 value=0.0721569 policy accuracy=55.1783% value accuracy=97.2806% mse=0.00508668
  202. step 123, policy=1.46764 value=0.0580823 policy accuracy=56.3452% value accuracy=97.8766% mse=0.00415759
  203. step 124, lr=0.02 policy=1.35991 value=0.0363863 mse=0.00245137 reg=0.165335 total=1.56163 (499.132 pos/s)
  204. step 124, policy=1.47554 value=0.150924 policy accuracy=55.3435% value accuracy=94.8017% mse=0.0101161
  205. step 125, policy=1.47015 value=0.0766632 policy accuracy=56.4553% value accuracy=97.0503% mse=0.00555999
  206. Model saved in file: ./networks/net-64x6/net-64x6-125
  207. saved as './networks/net-64x6/net-64x6-125' 5.93M
  208. Weights saved in file: ./networks/net-64x6/net-64x6-125
  209. step 126, lr=0.02 policy=1.40303 value=0.0316126 mse=0.0021891 reg=0.165346 total=1.59999 (448.779 pos/s)
  210. step 126, policy=1.44495 value=0.0301196 policy accuracy=56.4103% value accuracy=98.9633% mse=0.00202699
  211. step 127, policy=1.44209 value=0.058846 policy accuracy=56.7208% value accuracy=97.7414% mse=0.00424663
  212. step 128, lr=0.02 policy=1.38716 value=0.0249481 mse=0.00159752 reg=0.165357 total=1.57747 (510.195 pos/s)
  213. step 128, policy=1.45074 value=0.0687183 policy accuracy=56.851% value accuracy=97.6212% mse=0.00454727
  214. step 129, policy=1.42657 value=0.0389764 policy accuracy=56.9661% value accuracy=98.5677% mse=0.00269681
  215. step 130, lr=0.02 policy=1.40283 value=0.0332756 mse=0.00222664 reg=0.165367 total=1.60147 (518.092 pos/s)
  216. step 130, policy=1.39214 value=0.0438238 policy accuracy=57.7474% value accuracy=98.4075% mse=0.00307759
  217. step 131, policy=1.42548 value=0.0351392 policy accuracy=57.0162% value accuracy=98.7179% mse=0.00239353
  218. step 132, lr=0.02 policy=1.34801 value=0.0350446 mse=0.00232659 reg=0.165377 total=1.54843 (511.617 pos/s)
  219. step 132, policy=1.41188 value=0.0401959 policy accuracy=56.7057% value accuracy=98.5627% mse=0.00274738
  220. step 133, policy=1.43617 value=0.0597019 policy accuracy=57.2466% value accuracy=97.8215% mse=0.00415663
  221. step 134, lr=0.02 policy=1.36573 value=0.0345205 mse=0.00239942 reg=0.165387 total=1.56564 (505.135 pos/s)
  222. step 134, policy=1.41271 value=0.0386149 policy accuracy=56.9661% value accuracy=98.6579% mse=0.002589
  223. step 135, policy=1.42902 value=0.0293317 policy accuracy=57.6022% value accuracy=98.9383% mse=0.00195793
  224. step 136, lr=0.02 policy=1.38392 value=0.0374338 mse=0.00256384 reg=0.165397 total=1.58675 (509.23 pos/s)
  225. step 136, policy=1.40861 value=0.0794386 policy accuracy=56.1949% value accuracy=97.1004% mse=0.00551633
  226. step 137, policy=1.41757 value=0.0390304 policy accuracy=57.5771% value accuracy=98.5026% mse=0.00272362
  227. step 138, lr=0.02 policy=1.30443 value=0.0310932 mse=0.00208538 reg=0.165407 total=1.50093 (511.121 pos/s)
  228. step 138, policy=1.38548 value=0.0538053 policy accuracy=57.1264% value accuracy=97.9968% mse=0.00378246
  229. step 139, policy=1.36003 value=0.0299147 policy accuracy=58.0779% value accuracy=99.0034% mse=0.00195365
  230. step 140, lr=0.02 policy=1.30796 value=0.0387751 mse=0.00253067 reg=0.165416 total=1.51215 (512.673 pos/s)
  231. step 140, policy=1.37481 value=0.0462006 policy accuracy=57.2566% value accuracy=98.3574% mse=0.0031497
  232. step 141, policy=1.34181 value=0.0278991 policy accuracy=58.8742% value accuracy=99.0885% mse=0.00179989
  233. step 142, lr=0.02 policy=1.30794 value=0.0346485 mse=0.00222748 reg=0.165424 total=1.50801 (507.976 pos/s)
  234. step 142, policy=1.3423 value=0.0410742 policy accuracy=58.2732% value accuracy=98.5176% mse=0.00282269
  235. step 143, policy=1.3494 value=0.0708503 policy accuracy=58.153% value accuracy=97.3808% mse=0.00502972
  236. step 144, lr=0.02 policy=1.29795 value=0.0442626 mse=0.00303104 reg=0.165432 total=1.50765 (510.713 pos/s)
  237. step 144, policy=1.34422 value=0.0235734 policy accuracy=57.9227% value accuracy=99.1787% mse=0.00156816
  238. step 145, policy=1.32619 value=0.0509756 policy accuracy=59.3149% value accuracy=98.0369% mse=0.00364619
  239. step 146, lr=0.02 policy=1.24384 value=0.0318595 mse=0.00222595 reg=0.165439 total=1.44114 (504.399 pos/s)
  240. step 146, policy=1.34197 value=0.0292114 policy accuracy=59.0695% value accuracy=99.0134% mse=0.00197204
  241. step 147, policy=1.33957 value=0.0610788 policy accuracy=57.9026% value accuracy=97.8916% mse=0.00425653
  242. step 148, lr=0.02 policy=1.26862 value=0.0351032 mse=0.00241093 reg=0.165446 total=1.46917 (508.609 pos/s)
  243. step 148, policy=1.30954 value=0.0342377 policy accuracy=59.1546% value accuracy=98.8532% mse=0.00225666
  244. step 149, policy=1.31668 value=0.0334435 policy accuracy=58.759% value accuracy=98.8431% mse=0.0022129
  245. step 150, lr=0.02 policy=1.28058 value=0.0264532 mse=0.00185739 reg=0.165452 total=1.47249 (506.041 pos/s)
  246. step 150, policy=1.28686 value=0.0350505 policy accuracy=59.7055% value accuracy=98.7179% mse=0.00232779
  247. WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/tensorflow/python/training/saver.py:966: remove_checkpoint (from tensorflow.python.training.checkpoint_management) is deprecated and will be removed in a future version.
  248. Instructions for updating:
  249. Use standard file APIs to delete files with this prefix.
  250. Model saved in file: ./networks/net-64x6/net-64x6-150
  251. saved as './networks/net-64x6/net-64x6-150' 5.92M
  252. Weights saved in file: ./networks/net-64x6/net-64x6-150
  253. step 151, policy=1.29491 value=0.0312603 policy accuracy=59.1046% value accuracy=98.9133% mse=0.00207376
  254. step 152, lr=0.002 policy=1.23883 value=0.0274324 mse=0.00194613 reg=0.165458 total=1.43172 (443.476 pos/s)
  255. step 152, policy=1.31991 value=0.0294025 policy accuracy=58.2282% value accuracy=98.9984% mse=0.00189113
  256. step 153, policy=1.30092 value=0.0239827 policy accuracy=59.2698% value accuracy=99.1687% mse=0.00154804
  257. step 154, lr=0.002 policy=1.24546 value=0.0358067 mse=0.00223915 reg=0.165459 total=1.44672 (507.225 pos/s)
  258. step 154, policy=1.30073 value=0.0228434 policy accuracy=59.0194% value accuracy=99.2688% mse=0.00144461
  259. step 155, policy=1.27645 value=0.0223655 policy accuracy=59.9259% value accuracy=99.2788% mse=0.00142641
  260. step 156, lr=0.002 policy=1.20149 value=0.0318304 mse=0.00214225 reg=0.165459 total=1.39878 (505.105 pos/s)
  261. step 156, policy=1.30692 value=0.0248686 policy accuracy=58.8642% value accuracy=99.1536% mse=0.00163396
  262. step 157, policy=1.27702 value=0.0234639 policy accuracy=59.7706% value accuracy=99.1987% mse=0.00153491
  263. step 158, lr=0.002 policy=1.21313 value=0.0260921 mse=0.0018061 reg=0.16546 total=1.40468 (502.155 pos/s)
  264. step 158, policy=1.27041 value=0.0229424 policy accuracy=60.1613% value accuracy=99.2338% mse=0.00145471
  265. step 159, policy=1.28235 value=0.0231256 policy accuracy=59.6154% value accuracy=99.1837% mse=0.00148423
  266. step 160, lr=0.002 policy=1.23124 value=0.0272644 mse=0.00181203 reg=0.16546 total=1.42396 (507.436 pos/s)
  267. step 160, policy=1.26856 value=0.022 policy accuracy=59.5102% value accuracy=99.1887% mse=0.00145403
  268. step 161, policy=1.27518 value=0.0219191 policy accuracy=59.4952% value accuracy=99.2788% mse=0.00137383
  269. step 162, lr=0.002 policy=1.23258 value=0.030324 mse=0.001954 reg=0.16546 total=1.42836 (500.76 pos/s)
  270. step 162, policy=1.29302 value=0.0216266 policy accuracy=59.2648% value accuracy=99.2538% mse=0.00140438
  271. step 163, policy=1.25922 value=0.0197857 policy accuracy=60.2865% value accuracy=99.349% mse=0.00128051
  272. step 164, lr=0.002 policy=1.20136 value=0.0310863 mse=0.00207158 reg=0.165461 total=1.39791 (507.278 pos/s)
  273. step 164, policy=1.26625 value=0.0255161 policy accuracy=59.8357% value accuracy=99.0635% mse=0.00169991
  274. step 165, policy=1.29369 value=0.0233319 policy accuracy=59.4501% value accuracy=99.2188% mse=0.00154463
  275. step 166, lr=0.002 policy=1.22341 value=0.0231163 mse=0.00140389 reg=0.165461 total=1.41199 (511 pos/s)
  276. step 166, policy=1.266 value=0.0223934 policy accuracy=60.1713% value accuracy=99.2338% mse=0.00145835
  277. step 167, policy=1.2873 value=0.0221821 policy accuracy=59.2047% value accuracy=99.2638% mse=0.00145442
  278. step 168, lr=0.002 policy=1.21745 value=0.0231332 mse=0.0016042 reg=0.165461 total=1.40604 (511.256 pos/s)
  279. step 168, policy=1.25469 value=0.0230103 policy accuracy=60.2614% value accuracy=99.1987% mse=0.00154024
  280. step 169, policy=1.26653 value=0.0211704 policy accuracy=60.1763% value accuracy=99.2588% mse=0.00138298
  281. step 170, lr=0.002 policy=1.20136 value=0.0267201 mse=0.00172116 reg=0.165462 total=1.39354 (506.015 pos/s)
  282. step 170, policy=1.25431 value=0.0218131 policy accuracy=60.3115% value accuracy=99.2688% mse=0.00135235
  283. step 171, policy=1.27992 value=0.0216144 policy accuracy=59.2448% value accuracy=99.2538% mse=0.0014123
  284. step 172, lr=0.002 policy=1.24254 value=0.02701 mse=0.00181049 reg=0.165462 total=1.43501 (503.968 pos/s)
  285. step 172, policy=1.26159 value=0.0218001 policy accuracy=59.9008% value accuracy=99.2338% mse=0.00141836
  286. step 173, policy=1.25486 value=0.0189353 policy accuracy=60.1212% value accuracy=99.3289% mse=0.00119938
  287. step 174, lr=0.002 policy=1.20532 value=0.0345228 mse=0.00249641 reg=0.165462 total=1.4053 (509.174 pos/s)
  288. step 174, policy=1.27775 value=0.0230889 policy accuracy=59.4201% value accuracy=99.2238% mse=0.00146896
  289. step 175, policy=1.25936 value=0.0239784 policy accuracy=60.2564% value accuracy=99.1486% mse=0.00154679
  290. Model saved in file: ./networks/net-64x6/net-64x6-175
  291. saved as './networks/net-64x6/net-64x6-175' 5.92M
  292. Weights saved in file: ./networks/net-64x6/net-64x6-175
  293. step 176, lr=0.002 policy=1.24388 value=0.0244878 mse=0.00168949 reg=0.165463 total=1.43383 (441.23 pos/s)
  294. step 176, policy=1.2391 value=0.0251432 policy accuracy=60.6871% value accuracy=99.1086% mse=0.00169305
  295. step 177, policy=1.28133 value=0.0271143 policy accuracy=59.2748% value accuracy=99.0084% mse=0.00188893
  296. step 178, lr=0.002 policy=1.27408 value=0.0321288 mse=0.00212811 reg=0.165463 total=1.47167 (503.395 pos/s)
  297. step 178, policy=1.25604 value=0.0261062 policy accuracy=60.4517% value accuracy=99.0335% mse=0.00174615
  298. step 179, policy=1.29299 value=0.0226967 policy accuracy=58.8492% value accuracy=99.2037% mse=0.0014905
  299. step 180, lr=0.002 policy=1.2052 value=0.0347314 mse=0.00233071 reg=0.165463 total=1.40539 (509.709 pos/s)
  300. step 180, policy=1.26204 value=0.0206188 policy accuracy=59.6454% value accuracy=99.2839% mse=0.001324
  301. step 181, policy=1.27586 value=0.0223662 policy accuracy=59.8357% value accuracy=99.2538% mse=0.00141255
  302. step 182, lr=0.002 policy=1.22392 value=0.0243579 mse=0.00156234 reg=0.165463 total=1.41374 (510.131 pos/s)
  303. step 182, policy=1.24439 value=0.0202509 policy accuracy=60.612% value accuracy=99.2839% mse=0.00131934
  304. step 183, policy=1.27687 value=0.0233019 policy accuracy=59.2999% value accuracy=99.2538% mse=0.00142671
  305. step 184, lr=0.002 policy=1.18237 value=0.0237622 mse=0.00159683 reg=0.165463 total=1.37159 (507.371 pos/s)
  306. step 184, policy=1.26251 value=0.0215867 policy accuracy=60.1412% value accuracy=99.2538% mse=0.00142746
  307. step 185, policy=1.25666 value=0.0211663 policy accuracy=59.996% value accuracy=99.2788% mse=0.001408
  308. step 186, lr=0.002 policy=1.21975 value=0.0247802 mse=0.00167067 reg=0.165463 total=1.41 (494.327 pos/s)
  309. step 186, policy=1.24449 value=0.0239839 policy accuracy=60.2564% value accuracy=99.1787% mse=0.00158837
  310. step 187, policy=1.24959 value=0.0230222 policy accuracy=60.2815% value accuracy=99.1887% mse=0.00152485
  311. step 188, lr=0.002 policy=1.23167 value=0.0296527 mse=0.00205334 reg=0.165464 total=1.42678 (503.82 pos/s)
  312. step 188, policy=1.24189 value=0.0223659 policy accuracy=60.5719% value accuracy=99.2288% mse=0.00146495
  313. step 189, policy=1.26099 value=0.0213795 policy accuracy=59.9359% value accuracy=99.2538% mse=0.00141732
  314. step 190, lr=0.002 policy=1.16674 value=0.031686 mse=0.00213579 reg=0.165464 total=1.36389 (509.057 pos/s)
  315. step 190, policy=1.26306 value=0.0200687 policy accuracy=59.8407% value accuracy=99.344% mse=0.00129129
  316. step 191, policy=1.27507 value=0.0210825 policy accuracy=59.6054% value accuracy=99.2338% mse=0.00139635
  317. step 192, lr=0.002 policy=1.22421 value=0.0298084 mse=0.00205099 reg=0.165464 total=1.41948 (501.447 pos/s)
  318. step 192, policy=1.27242 value=0.0225102 policy accuracy=59.2598% value accuracy=99.1887% mse=0.00151096
  319. step 193, policy=1.25599 value=0.019369 policy accuracy=60.4667% value accuracy=99.2989% mse=0.00126998
  320. step 194, lr=0.002 policy=1.14039 value=0.0266224 mse=0.00176909 reg=0.165464 total=1.33248 (504.672 pos/s)
  321. step 194, policy=1.25173 value=0.0209325 policy accuracy=60.1713% value accuracy=99.3289% mse=0.00133916
  322. step 195, policy=1.26184 value=0.0183309 policy accuracy=59.9459% value accuracy=99.359% mse=0.00118148
  323. step 196, lr=0.002 policy=1.19487 value=0.0210491 mse=0.00142485 reg=0.165464 total=1.38138 (505.158 pos/s)
  324. step 196, policy=1.26941 value=0.0210349 policy accuracy=59.7005% value accuracy=99.2588% mse=0.00138249
  325. step 197, policy=1.23278 value=0.0203591 policy accuracy=60.6621% value accuracy=99.2839% mse=0.00132002
  326. step 198, lr=0.002 policy=1.21837 value=0.0297733 mse=0.0018975 reg=0.165464 total=1.41361 (498.515 pos/s)
  327. step 198, policy=1.2345 value=0.0212084 policy accuracy=60.6721% value accuracy=99.2738% mse=0.00141741
  328. step 199, policy=1.2604 value=0.0200773 policy accuracy=59.6755% value accuracy=99.3389% mse=0.00129989
  329. step 200, lr=0.002 policy=1.2133 value=0.0269627 mse=0.00183617 reg=0.165464 total=1.40573 (501.996 pos/s)
  330. step 200, policy=1.25323 value=0.0201518 policy accuracy=59.9058% value accuracy=99.2738% mse=0.00133076
  331. Model saved in file: ./networks/net-64x6/net-64x6-200
  332. saved as './networks/net-64x6/net-64x6-200' 5.92M
  333. Weights saved in file: ./networks/net-64x6/net-64x6-200
  334. step 201, policy=1.25184 value=0.0206857 policy accuracy=60.3415% value accuracy=99.3339% mse=0.00132114
  335. step 202, lr=0.001 policy=1.17172 value=0.0218427 mse=0.00142379 reg=0.165464 total=1.35903 (443.064 pos/s)
  336. step 202, policy=1.2378 value=0.0170951 policy accuracy=60.5919% value accuracy=99.4141% mse=0.00112384
  337. step 203, policy=1.25191 value=0.0231056 policy accuracy=60.6621% value accuracy=99.2087% mse=0.00148458
  338. step 204, lr=0.001 policy=1.1859 value=0.0253925 mse=0.00177213 reg=0.165465 total=1.37676 (502.108 pos/s)
  339. step 204, policy=1.24939 value=0.0208258 policy accuracy=59.7756% value accuracy=99.3039% mse=0.00137109
  340. step 205, policy=1.25584 value=0.0187831 policy accuracy=59.981% value accuracy=99.3239% mse=0.00127965
  341. step 206, lr=0.001 policy=1.21741 value=0.0288127 mse=0.00174453 reg=0.165465 total=1.41168 (506.477 pos/s)
  342. step 206, policy=1.27255 value=0.0210646 policy accuracy=59.5553% value accuracy=99.2638% mse=0.00138507
  343. step 207, policy=1.24796 value=0.0219682 policy accuracy=60.5319% value accuracy=99.2188% mse=0.00141653
  344. step 208, lr=0.001 policy=1.18807 value=0.0215104 mse=0.00146637 reg=0.165465 total=1.37504 (503.905 pos/s)
  345. step 208, policy=1.24351 value=0.0176754 policy accuracy=60.3065% value accuracy=99.399% mse=0.00116096
  346. step 209, policy=1.25396 value=0.0177605 policy accuracy=59.9459% value accuracy=99.384% mse=0.00115923
  347. step 210, lr=0.001 policy=1.19321 value=0.0249083 mse=0.00171246 reg=0.165465 total=1.38358 (496.386 pos/s)
  348. step 210, policy=1.25941 value=0.0205841 policy accuracy=59.981% value accuracy=99.2989% mse=0.00133351
  349. step 211, policy=1.24808 value=0.0228531 policy accuracy=60.6671% value accuracy=99.1536% mse=0.00149428
  350. step 212, lr=0.001 policy=1.19239 value=0.0252935 mse=0.00166547 reg=0.165465 total=1.38315 (504.885 pos/s)
  351. step 212, policy=1.27131 value=0.0204835 policy accuracy=59.4601% value accuracy=99.3139% mse=0.00128876
  352. step 213, policy=1.24736 value=0.018849 policy accuracy=60.3966% value accuracy=99.399% mse=0.0011987
  353. step 214, lr=0.001 policy=1.17573 value=0.0307521 mse=0.00192719 reg=0.165465 total=1.37195 (498.898 pos/s)
  354. step 214, policy=1.22095 value=0.0191281 policy accuracy=61.228% value accuracy=99.2889% mse=0.00130285
  355. step 215, policy=1.25639 value=0.0200687 policy accuracy=59.7907% value accuracy=99.2738% mse=0.00129683
  356. step 216, lr=0.001 policy=1.25258 value=0.035544 mse=0.00249965 reg=0.165465 total=1.45359 (499.986 pos/s)
  357. step 216, policy=1.23984 value=0.0229538 policy accuracy=60.5419% value accuracy=99.1787% mse=0.00154103
  358. step 217, policy=1.24438 value=0.0214958 policy accuracy=60.5469% value accuracy=99.2638% mse=0.00139993
  359. step 218, lr=0.001 policy=1.23048 value=0.0308985 mse=0.00213316 reg=0.165465 total=1.42684 (506.175 pos/s)
  360. step 218, policy=1.23483 value=0.022774 policy accuracy=61.1979% value accuracy=99.1336% mse=0.00154141
  361. step 219, policy=1.24735 value=0.0232727 policy accuracy=60.0911% value accuracy=99.1937% mse=0.00154187
  362. step 220, lr=0.001 policy=1.16012 value=0.02455 mse=0.00155945 reg=0.165465 total=1.35014 (504.458 pos/s)
  363. step 220, policy=1.25397 value=0.0197142 policy accuracy=60.4067% value accuracy=99.2688% mse=0.00129151
  364. step 221, policy=1.22792 value=0.023291 policy accuracy=60.5218% value accuracy=99.2037% mse=0.00150747
  365. step 222, lr=0.001 policy=1.20187 value=0.023696 mse=0.00152106 reg=0.165465 total=1.39104 (500.192 pos/s)
  366. step 222, policy=1.25437 value=0.0197265 policy accuracy=60.031% value accuracy=99.2889% mse=0.00129433
  367. step 223, policy=1.2283 value=0.0194742 policy accuracy=60.9625% value accuracy=99.3339% mse=0.00122529
  368. step 224, lr=0.001 policy=1.22335 value=0.027128 mse=0.00173456 reg=0.165465 total=1.41594 (504.013 pos/s)
  369. step 224, policy=1.25251 value=0.0182498 policy accuracy=60.2264% value accuracy=99.364% mse=0.00119649
  370. step 225, policy=1.22069 value=0.019574 policy accuracy=60.9525% value accuracy=99.2939% mse=0.00129106
  371. Model saved in file: ./networks/net-64x6/net-64x6-225
  372. saved as './networks/net-64x6/net-64x6-225' 5.92M
  373. Weights saved in file: ./networks/net-64x6/net-64x6-225
  374. step 226, lr=0.001 policy=1.17395 value=0.0272705 mse=0.00188056 reg=0.165465 total=1.36669 (436.238 pos/s)
  375. step 226, policy=1.2539 value=0.0210144 policy accuracy=59.9509% value accuracy=99.2388% mse=0.00138496
  376. step 227, policy=1.25606 value=0.0204532 policy accuracy=60.1512% value accuracy=99.3089% mse=0.0012914
  377. step 228, lr=0.001 policy=1.15661 value=0.023755 mse=0.00162011 reg=0.165465 total=1.34583 (507.5 pos/s)
  378. step 228, policy=1.24386 value=0.0204902 policy accuracy=60.2764% value accuracy=99.2738% mse=0.0013424
  379. step 229, policy=1.25283 value=0.0218524 policy accuracy=60.5218% value accuracy=99.2438% mse=0.00143679
  380. step 230, lr=0.001 policy=1.18685 value=0.0245462 mse=0.00163358 reg=0.165465 total=1.37686 (500.093 pos/s)
  381. step 230, policy=1.25082 value=0.0225445 policy accuracy=59.7957% value accuracy=99.2087% mse=0.00151216
  382. step 231, policy=1.23528 value=0.0205332 policy accuracy=60.1913% value accuracy=99.3039% mse=0.00132025
  383. step 232, lr=0.001 policy=1.18451 value=0.0263748 mse=0.00175596 reg=0.165465 total=1.37635 (500.627 pos/s)
  384. step 232, policy=1.2379 value=0.0198594 policy accuracy=60.5619% value accuracy=99.369% mse=0.00123103
  385. step 233, policy=1.25348 value=0.0204912 policy accuracy=59.9309% value accuracy=99.3189% mse=0.00129914
  386. step 234, lr=0.001 policy=1.18261 value=0.0261637 mse=0.00172371 reg=0.165465 total=1.37424 (502.417 pos/s)
  387. step 234, policy=1.24209 value=0.0193593 policy accuracy=60.5419% value accuracy=99.349% mse=0.00124185
  388. step 235, policy=1.23894 value=0.0171413 policy accuracy=60.5118% value accuracy=99.4141% mse=0.00113147
  389. step 236, lr=0.001 policy=1.227 value=0.0282658 mse=0.00189406 reg=0.165465 total=1.42073 (498.04 pos/s)
  390. step 236, policy=1.24556 value=0.0186377 policy accuracy=60.612% value accuracy=99.2939% mse=0.00124456
  391. step 237, policy=1.22983 value=0.0218438 policy accuracy=60.9475% value accuracy=99.2087% mse=0.00140678
  392. step 238, lr=0.001 policy=1.1743 value=0.0242486 mse=0.00159003 reg=0.165465 total=1.36401 (496.974 pos/s)
  393. step 238, policy=1.25518 value=0.0187933 policy accuracy=59.7306% value accuracy=99.3039% mse=0.00124864
  394. step 239, policy=1.24689 value=0.0205995 policy accuracy=60.2464% value accuracy=99.2388% mse=0.0013814
  395. step 240, lr=0.001 policy=1.16418 value=0.0237015 mse=0.00157006 reg=0.165466 total=1.35335 (502.907 pos/s)
  396. step 240, policy=1.23805 value=0.0198671 policy accuracy=61.0327% value accuracy=99.3139% mse=0.00127863
  397. step 241, policy=1.2389 value=0.019889 policy accuracy=60.5869% value accuracy=99.2438% mse=0.00134128
  398. step 242, lr=0.001 policy=1.18005 value=0.0259543 mse=0.00174123 reg=0.165466 total=1.37146 (507.909 pos/s)
  399. step 242, policy=1.22669 value=0.0201883 policy accuracy=61.243% value accuracy=99.3189% mse=0.0012725
  400. step 243, policy=1.25303 value=0.0194043 policy accuracy=59.9008% value accuracy=99.3289% mse=0.00126355
  401. step 244, lr=0.001 policy=1.17241 value=0.0227854 mse=0.00145794 reg=0.165466 total=1.36066 (502.148 pos/s)
  402. step 244, policy=1.23134 value=0.0208436 policy accuracy=60.3816% value accuracy=99.2288% mse=0.00138648
  403. step 245, policy=1.23316 value=0.0183569 policy accuracy=60.8073% value accuracy=99.3389% mse=0.00124607
  404. step 246, lr=0.001 policy=1.13241 value=0.0251316 mse=0.00171253 reg=0.165466 total=1.32301 (502.399 pos/s)
  405. step 246, policy=1.23374 value=0.0213396 policy accuracy=60.6721% value accuracy=99.2989% mse=0.00135768
  406. step 247, policy=1.24326 value=0.0202389 policy accuracy=60.4417% value accuracy=99.2087% mse=0.00139971
  407. step 248, lr=0.001 policy=1.20973 value=0.0210593 mse=0.00139744 reg=0.165466 total=1.39626 (502.639 pos/s)
  408. step 248, policy=1.23577 value=0.0199828 policy accuracy=60.8023% value accuracy=99.3189% mse=0.00128317
  409. step 249, policy=1.23878 value=0.0193874 policy accuracy=60.2013% value accuracy=99.2788% mse=0.00130918
  410. step 250, lr=0.001 policy=1.17329 value=0.026008 mse=0.00168977 reg=0.165466 total=1.36477 (499.686 pos/s)
  411. step 250, policy=1.25581 value=0.0220524 policy accuracy=59.7256% value accuracy=99.2538% mse=0.00142116
  412. Model saved in file: ./networks/net-64x6/net-64x6-250
  413. saved as './networks/net-64x6/net-64x6-250' 5.92M
  414. Weights saved in file: ./networks/net-64x6/net-64x6-250
  415. step 251, policy=1.23532 value=0.0199454 policy accuracy=60.6721% value accuracy=99.3139% mse=0.00127187
  416. step 252, lr=0.001 policy=1.18025 value=0.0308518 mse=0.00209685 reg=0.165466 total=1.37657 (437.222 pos/s)
  417. step 252, policy=1.24451 value=0.0186652 policy accuracy=60.622% value accuracy=99.3039% mse=0.00123752
  418. step 253, policy=1.24919 value=0.0207394 policy accuracy=60.1913% value accuracy=99.2989% mse=0.00136858
  419. step 254, lr=0.001 policy=1.16767 value=0.0241271 mse=0.001646 reg=0.165466 total=1.35726 (497.275 pos/s)
  420. step 254, policy=1.22484 value=0.0210473 policy accuracy=60.5369% value accuracy=99.2788% mse=0.00141528
  421. step 255, policy=1.24031 value=0.021349 policy accuracy=60.1512% value accuracy=99.2588% mse=0.00142061
  422. step 256, lr=0.001 policy=1.19393 value=0.0252226 mse=0.00168964 reg=0.165466 total=1.38462 (500.585 pos/s)
  423. step 256, policy=1.23594 value=0.0200846 policy accuracy=60.5919% value accuracy=99.369% mse=0.00124897
  424. step 257, policy=1.2313 value=0.0191049 policy accuracy=60.4718% value accuracy=99.404% mse=0.00118744
  425. step 258, lr=0.001 policy=1.15702 value=0.0228179 mse=0.00148703 reg=0.165466 total=1.3453 (506.653 pos/s)
  426. step 258, policy=1.22304 value=0.0204863 policy accuracy=60.7422% value accuracy=99.359% mse=0.00127934
  427. step 259, policy=1.23839 value=0.0198199 policy accuracy=60.5819% value accuracy=99.2688% mse=0.00137233
  428. step 260, lr=0.001 policy=1.20317 value=0.0299077 mse=0.00206603 reg=0.165466 total=1.39854 (505.763 pos/s)
  429. step 260, policy=1.22614 value=0.0196121 policy accuracy=60.7622% value accuracy=99.374% mse=0.00120375
  430. step 261, policy=1.24833 value=0.0175247 policy accuracy=60.1312% value accuracy=99.384% mse=0.00115127
  431. step 262, lr=0.001 policy=1.17161 value=0.0272812 mse=0.00181578 reg=0.165466 total=1.36436 (499.937 pos/s)
  432. step 262, policy=1.22469 value=0.0183 policy accuracy=60.6721% value accuracy=99.3389% mse=0.00121525
  433. step 263, policy=1.23753 value=0.0195023 policy accuracy=60.4667% value accuracy=99.3089% mse=0.00127146
  434. step 264, lr=0.001 policy=1.18505 value=0.0211663 mse=0.00137527 reg=0.165466 total=1.37168 (498.372 pos/s)
  435. step 264, policy=1.24299 value=0.0180063 policy accuracy=60.622% value accuracy=99.4391% mse=0.00110316
  436. step 265, policy=1.22216 value=0.0185536 policy accuracy=60.7272% value accuracy=99.2939% mse=0.00121275
  437. step 266, lr=0.001 policy=1.17115 value=0.0283745 mse=0.00179509 reg=0.165466 total=1.36499 (496.322 pos/s)
  438. step 266, policy=1.23751 value=0.0215052 policy accuracy=60.3365% value accuracy=99.2288% mse=0.00140165
  439. step 267, policy=1.23802 value=0.0184837 policy accuracy=60.1562% value accuracy=99.359% mse=0.00119979
  440. step 268, lr=0.001 policy=1.17055 value=0.0266437 mse=0.00177464 reg=0.165466 total=1.36266 (496.836 pos/s)
  441. step 268, policy=1.21478 value=0.020108 policy accuracy=61.233% value accuracy=99.2438% mse=0.00136724
  442. step 269, policy=1.22708 value=0.0204729 policy accuracy=60.5519% value accuracy=99.2488% mse=0.00135561
  443. step 270, lr=0.001 policy=1.17088 value=0.0248705 mse=0.00162734 reg=0.165466 total=1.36121 (484.403 pos/s)
  444. step 270, policy=1.22379 value=0.0189886 policy accuracy=60.9625% value accuracy=99.3139% mse=0.00126566
Advertisement
Add Comment
Please, Sign In to add comment
Advertisement