log.txt

learning_rate = 0.001

episode: 0, epsilon: 0.5, loss: 0, win_rate: [ 0.  0.  1.]
episode: 2000, epsilon: 0.5250899999998858, loss: 0.2945, win_rate: [ 0.09445277  0.46676662  0.43878061]
episode: 4000, epsilon: 0.555159999999749, loss: 0.0483606, win_rate: [ 0.11897026  0.45188703  0.42914271]
episode: 6000, epsilon: 0.5846699999996147, loss: 0.0617657, win_rate: [ 0.12631228  0.44009332  0.4335944 ]
episode: 8000, epsilon: 0.6143199999994797, loss: 0.0887938, win_rate: [ 0.13773278  0.43232096  0.42994626]
episode: 10000, epsilon: 0.6440799999993443, loss: 0.0697613, win_rate: [ 0.1469853   0.4309569   0.42205779]
episode: 12000, epsilon: 0.6736599999992097, loss: 0.0504912, win_rate: [ 0.15507041  0.42479793  0.42013166]
episode: 14000, epsilon: 0.703459999999074, loss: 0.0890487, win_rate: [ 0.16213128  0.41975573  0.41811299]
episode: 16000, epsilon: 0.7327599999989407, loss: 0.0360363, win_rate: [ 0.16648959  0.41816136  0.41534904]
episode: 18000, epsilon: 0.7621099999988071, loss: 0.0938439, win_rate: [ 0.17132382  0.41331037  0.41536581]
episode: 20000, epsilon: 0.7914399999986736, loss: 0.0530558, win_rate: [ 0.17579121  0.41072946  0.41347933]
episode: 22000, epsilon: 0.8209299999985394, loss: 0.162449, win_rate: [ 0.18085542  0.40843598  0.4107086 ]
episode: 24000, epsilon: 0.8503999999984053, loss: 0.0342156, win_rate: [ 0.18653389  0.4054831   0.407983  ]
episode: 26000, epsilon: 0.879479999998273, loss: 0.104467, win_rate: [ 0.19176186  0.40225376  0.40598439]
episode: 28000, epsilon: 0.9, loss: 0.0211573, win_rate: [ 0.19745723  0.39859291  0.40394986]
episode: 30000, epsilon: 0.9, loss: 0.0851524, win_rate: [ 0.20195993  0.39672011  0.40131996]
episode: 32000, epsilon: 0.9, loss: 0.106228, win_rate: [ 0.20621231  0.39548764  0.39830005]
episode: 34000, epsilon: 0.9, loss: 0.0472208, win_rate: [ 0.20940561  0.39425311  0.39634128]
episode: 36000, epsilon: 0.9, loss: 0.0424313, win_rate: [ 0.21191078  0.39298908  0.39510014]
episode: 38000, epsilon: 0.9, loss: 0.137321, win_rate: [ 0.21486277  0.39088445  0.39425278]
episode: 40000, epsilon: 0.9, loss: 0.086276, win_rate: [ 0.21736957  0.38914027  0.39349016]
episode: 42000, epsilon: 0.9, loss: 0.0411554, win_rate: [ 0.21997095  0.38806219  0.39196686]
episode: 44000, epsilon: 0.9, loss: 0.0811864, win_rate: [ 0.2226313   0.38635486  0.39101384]
episode: 46000, epsilon: 0.9, loss: 0.0357934, win_rate: [ 0.22523423  0.38520902  0.38955675]
episode: 48000, epsilon: 0.9, loss: 0.0882131, win_rate: [ 0.22745359  0.38330451  0.38924189]
episode: 50000, epsilon: 0.9, loss: 0.0540792, win_rate: [ 0.22907542  0.38265235  0.38827223]
episode: 52000, epsilon: 0.9, loss: 0.0430649, win_rate: [ 0.22988019  0.38168497  0.38843484]
episode: 54000, epsilon: 0.9, loss: 0.0392573, win_rate: [ 0.23182904  0.38003     0.38814096]
episode: 56000, epsilon: 0.9, loss: 0.0997899, win_rate: [ 0.23404939  0.37901109  0.38693952]
episode: 58000, epsilon: 0.9, loss: 0.0652466, win_rate: [ 0.23599593  0.3778728   0.38613127]
episode: 60000, epsilon: 0.9, loss: 0.26159, win_rate: [ 0.23766271  0.37667706  0.38566024]
episode: 62000, epsilon: 0.9, loss: 0.0388225, win_rate: [ 0.23954452  0.37599394  0.38446154]
episode: 64000, epsilon: 0.9, loss: 0.0423146, win_rate: [ 0.2414181   0.37518164  0.38340026]
episode: 66000, epsilon: 0.9, loss: 0.0677372, win_rate: [ 0.24328419  0.37388827  0.38282753]
episode: 68000, epsilon: 0.9, loss: 0.0319142, win_rate: [ 0.24561404  0.37289158  0.38149439]
episode: 70000, epsilon: 0.9, loss: 0.0202877, win_rate: [ 0.2476536   0.37175183  0.38059456]
episode: 72000, epsilon: 0.9, loss: 0.0463033, win_rate: [ 0.24962153  0.37060596  0.3797725 ]
episode: 74000, epsilon: 0.9, loss: 0.249103, win_rate: [ 0.25090201  0.37013013  0.37896785]
episode: 76000, epsilon: 0.9, loss: 0.0254147, win_rate: [ 0.25229931  0.36931093  0.37838976]
episode: 78000, epsilon: 0.9, loss: 0.0734374, win_rate: [ 0.25362495  0.36879014  0.3775849 ]
episode: 80000, epsilon: 0.9, loss: 0.0659173, win_rate: [ 0.25448432  0.36878289  0.37673279]
episode: 82000, epsilon: 0.9, loss: 0.0234815, win_rate: [ 0.25548469  0.36789795  0.37661736]
episode: 84000, epsilon: 0.9, loss: 0.0362075, win_rate: [ 0.25672313  0.36747182  0.37580505]
episode: 86000, epsilon: 0.9, loss: 0.159669, win_rate: [ 0.25755514  0.36656551  0.37587935]
episode: 88000, epsilon: 0.9, loss: 0.0226907, win_rate: [ 0.25881524  0.36621175  0.37497301]
episode: 90000, epsilon: 0.9, loss: 0.03181, win_rate: [ 0.26007489  0.36567371  0.3742514 ]
episode: 92000, epsilon: 0.9, loss: 0.0713771, win_rate: [ 0.26102977  0.36521342  0.37375681]
episode: 94000, epsilon: 0.9, loss: 0.0475698, win_rate: [ 0.261827    0.36494293  0.37323007]
episode: 96000, epsilon: 0.9, loss: 0.131037, win_rate: [ 0.26272643  0.3643712   0.37290237]
episode: 98000, epsilon: 0.9, loss: 0.0853201, win_rate: [ 0.26351772  0.36355751  0.37292477]
episode: 100000, epsilon: 0.9, loss: 0.0277419, win_rate: [ 0.26419736  0.36332637  0.37247628]
episode: 102000, epsilon: 0.9, loss: 0.0351002, win_rate: [ 0.26505622  0.3631533   0.37179047]
episode: 104000, epsilon: 0.9, loss: 0.0918793, win_rate: [ 0.26575706  0.3626119   0.37163104]
episode: 106000, epsilon: 0.9, loss: 0.0439326, win_rate: [ 0.26678994  0.36208149  0.37112857]
episode: 108000, epsilon: 0.9, loss: 0.115934, win_rate: [ 0.26780308  0.36199665  0.37020028]
episode: 110000, epsilon: 0.9, loss: 0.0483172, win_rate: [ 0.26868847  0.3615058   0.36980573]
episode: 112000, epsilon: 0.9, loss: 0.0893005, win_rate: [ 0.26972973  0.36096106  0.3693092 ]
episode: 114000, epsilon: 0.9, loss: 0.102788, win_rate: [ 0.27062043  0.3603828   0.36899676]
episode: 116000, epsilon: 0.9, loss: 0.10074, win_rate: [ 0.27141145  0.36022103  0.36836751]
episode: 118000, epsilon: 0.9, loss: 0.0491942, win_rate: [ 0.2722943   0.35964102  0.36806468]
episode: 120000, epsilon: 0.9, loss: 0.0595602, win_rate: [ 0.27303939  0.35943034  0.36753027]
episode: 122000, epsilon: 0.9, loss: 0.0489912, win_rate: [ 0.27376825  0.35934132  0.36689044]
episode: 124000, epsilon: 0.9, loss: 0.0648632, win_rate: [ 0.27445746  0.35885194  0.36669059]
episode: 126000, epsilon: 0.9, loss: 0.05712, win_rate: [ 0.27527559  0.35842573  0.36629868]
episode: 128000, epsilon: 0.9, loss: 0.0556854, win_rate: [ 0.27605253  0.3583097   0.36563777]
episode: 130000, epsilon: 0.9, loss: 0.0302121, win_rate: [ 0.27698248  0.35775109  0.36526642]
episode: 132000, epsilon: 0.9, loss: 0.0261284, win_rate: [ 0.27745244  0.35743669  0.36511087]
episode: 134000, epsilon: 0.9, loss: 0.0359763, win_rate: [ 0.277983    0.35711674  0.36490026]
episode: 136000, epsilon: 0.9, loss: 0.0414283, win_rate: [ 0.27847589  0.35687973  0.36464438]
episode: 138000, epsilon: 0.9, loss: 0.137298, win_rate: [ 0.27898349  0.35661336  0.36440316]
episode: 140000, epsilon: 0.9, loss: 0.0483653, win_rate: [ 0.27941943  0.35636888  0.36421168]
episode: 142000, epsilon: 0.9, loss: 0.0606391, win_rate: [ 0.28004732  0.35611017  0.36384251]
episode: 144000, epsilon: 0.9, loss: 0.0778435, win_rate: [ 0.28051889  0.35593503  0.36354609]
episode: 146000, epsilon: 0.9, loss: 0.192776, win_rate: [ 0.28094328  0.35594277  0.36311395]
episode: 148000, epsilon: 0.9, loss: 0.0472505, win_rate: [ 0.28153188  0.35555841  0.36290971]
episode: 150000, epsilon: 0.9, loss: 0.0475737, win_rate: [ 0.28223145  0.35517763  0.36259092]
episode: 152000, epsilon: 0.9, loss: 0.0321817, win_rate: [ 0.28243893  0.35519503  0.36236604]
episode: 154000, epsilon: 0.9, loss: 0.14176, win_rate: [ 0.28317349  0.35461458  0.36221193]
episode: 156000, epsilon: 0.9, loss: 0.0289719, win_rate: [ 0.2836328   0.35436311  0.36200409]
episode: 158000, epsilon: 0.9, loss: 0.0755346, win_rate: [ 0.28411845  0.35403573  0.36184581]
episode: 160000, epsilon: 0.9, loss: 0.0690059, win_rate: [ 0.28456072  0.35392279  0.36151649]
episode: 162000, epsilon: 0.9, loss: 0.0761212, win_rate: [ 0.28487478  0.35386819  0.36125703]
episode: 164000, epsilon: 0.9, loss: 0.0786304, win_rate: [ 0.28544948  0.35356492  0.3609856 ]
episode: 166000, epsilon: 0.9, loss: 0.152283, win_rate: [ 0.28585972  0.35342558  0.36071469]
episode: 168000, epsilon: 0.9, loss: 0.0375376, win_rate: [ 0.28656972  0.35289671  0.36053357]
episode: 170000, epsilon: 0.9, loss: 0.0694601, win_rate: [ 0.28709243  0.35255087  0.3603567 ]
episode: 172000, epsilon: 0.9, loss: 0.0662952, win_rate: [ 0.28761461  0.35227702  0.36010837]
episode: 174000, epsilon: 0.9, loss: 0.126031, win_rate: [ 0.28794087  0.35187154  0.36018759]
episode: 176000, epsilon: 0.9, loss: 0.0441226, win_rate: [ 0.287987    0.35173664  0.36027636]
episode: 178000, epsilon: 0.9, loss: 0.0939541, win_rate: [ 0.28838602  0.3516778   0.35993618]
episode: 180000, epsilon: 0.9, loss: 0.0806355, win_rate: [ 0.28870951  0.35121472  0.36007578]
episode: 182000, epsilon: 0.9, loss: 0.225121, win_rate: [ 0.28930061  0.35102005  0.35967934]
episode: 184000, epsilon: 0.9, loss: 0.0905873, win_rate: [ 0.28973756  0.3508894   0.35937305]
episode: 186000, epsilon: 0.9, loss: 0.0462258, win_rate: [ 0.29015435  0.35055188  0.35929377]
episode: 188000, epsilon: 0.9, loss: 0.0449081, win_rate: [ 0.29040271  0.35033856  0.35925873]
episode: 190000, epsilon: 0.9, loss: 0.0201836, win_rate: [ 0.29060373  0.35022447  0.35917179]
episode: 192000, epsilon: 0.9, loss: 0.0239741, win_rate: [ 0.29076932  0.3501388   0.35909188]
episode: 194000, epsilon: 0.9, loss: 0.0344469, win_rate: [ 0.29091087  0.35013222  0.35895691]
episode: 196000, epsilon: 0.9, loss: 0.0569738, win_rate: [ 0.29098321  0.35012576  0.35889103]
episode: 198000, epsilon: 0.9, loss: 0.0272727, win_rate: [ 0.2912359  0.3501346  0.3586295]
episode: 200000, epsilon: 0.9, loss: 0.0319101, win_rate: [ 0.29155354  0.34988325  0.35856321]
episode: 202000, epsilon: 0.9, loss: 0.0657818, win_rate: [ 0.29199855  0.34967649  0.35832496]
episode: 204000, epsilon: 0.9, loss: 0.0822009, win_rate: [ 0.29224367  0.34940515  0.35835118]
episode: 206000, epsilon: 0.9, loss: 0.0827054, win_rate: [ 0.29271703  0.34924102  0.35804195]
episode: 208000, epsilon: 0.9, loss: 0.0876145, win_rate: [ 0.29290244  0.34921467  0.35788289]
episode: 210000, epsilon: 0.9, loss: 0.0300222, win_rate: [ 0.29317003  0.34905548  0.35777449]
episode: 212000, epsilon: 0.9, loss: 0.0452345, win_rate: [ 0.2934373   0.34886628  0.35769643]
episode: 214000, epsilon: 0.9, loss: 0.086108, win_rate: [ 0.29379302  0.34877874  0.35742824]
episode: 216000, epsilon: 0.9, loss: 0.130069, win_rate: [ 0.29425327  0.34847987  0.35726686]
episode: 218000, epsilon: 0.9, loss: 0.0776821, win_rate: [ 0.29454452  0.34826904  0.35718643]
episode: 220000, epsilon: 0.9, loss: 0.0832973, win_rate: [ 0.29468048  0.34832569  0.35699383]
episode: 222000, epsilon: 0.9, loss: 0.0427337, win_rate: [ 0.2950302   0.34815158  0.35681821]
episode: 224000, epsilon: 0.9, loss: 0.0352238, win_rate: [ 0.29522636  0.34787791  0.35689573]
episode: 226000, epsilon: 0.9, loss: 0.0466583, win_rate: [ 0.29551197  0.34767988  0.35680816]
episode: 228000, epsilon: 0.9, loss: 0.0672912, win_rate: [ 0.29570484  0.3476169   0.35667826]
episode: 230000, epsilon: 0.9, loss: 0.0366899, win_rate: [ 0.29593784  0.34758545  0.35647671]
episode: 232000, epsilon: 0.9, loss: 0.116823, win_rate: [ 0.29611941  0.34756747  0.35631312]
episode: 234000, epsilon: 0.9, loss: 0.0342191, win_rate: [ 0.29646027  0.34742159  0.35611814]
episode: 236000, epsilon: 0.9, loss: 0.254224, win_rate: [ 0.29668095  0.34732056  0.35599849]
episode: 238000, epsilon: 0.9, loss: 0.104247, win_rate: [ 0.29704077  0.34718342  0.35577582]
episode: 240000, epsilon: 0.9, loss: 0.0652708, win_rate: [ 0.29731126  0.34704855  0.35564018]
episode: 242000, epsilon: 0.9, loss: 0.132359, win_rate: [ 0.29762274  0.34680022  0.35557704]
episode: 244000, epsilon: 0.9, loss: 0.043645, win_rate: [ 0.29807665  0.34653547  0.35538789]
episode: 246000, epsilon: 0.9, loss: 0.056568, win_rate: [ 0.29840936  0.34641729  0.35517335]
episode: 248000, epsilon: 0.9, loss: 0.0308357, win_rate: [ 0.29852299  0.3461478   0.35532921]
episode: 250000, epsilon: 0.9, loss: 0.122293, win_rate: [ 0.29872281  0.34604262  0.35523458]
episode: 252000, epsilon: 0.9, loss: 0.104931, win_rate: [ 0.29897104  0.34582006  0.35520891]
episode: 254000, epsilon: 0.9, loss: 0.0407437, win_rate: [ 0.29919174  0.34562462  0.35518364]
episode: 256000, epsilon: 0.9, loss: 0.0428849, win_rate: [ 0.29919024  0.34560412  0.35520564]
episode: 258000, epsilon: 0.9, loss: 0.0982826, win_rate: [ 0.2992469   0.34556455  0.35518855]
episode: 260000, epsilon: 0.9, loss: 0.105309, win_rate: [ 0.29932962  0.34557559  0.35509479]
episode: 262000, epsilon: 0.9, loss: 0.0842547, win_rate: [ 0.29941489  0.34548723  0.35509788]
episode: 264000, epsilon: 0.9, loss: 0.0204638, win_rate: [ 0.29973371  0.34528278  0.3549835 ]
episode: 266000, epsilon: 0.9, loss: 0.0500088, win_rate: [ 0.29977331  0.34524682  0.35497987]
episode: 268000, epsilon: 0.9, loss: 0.0835725, win_rate: [ 0.29993545  0.34525244  0.35481211]
episode: 270000, epsilon: 0.9, loss: 0.108896, win_rate: [ 0.30027296  0.34519131  0.35453572]
episode: 272000, epsilon: 0.9, loss: 0.0243984, win_rate: [ 0.30034081  0.34516785  0.35449134]
episode: 274000, epsilon: 0.9, loss: 0.0340141, win_rate: [ 0.30054635  0.34498779  0.35446586]
episode: 276000, epsilon: 0.9, loss: 0.059728, win_rate: [ 0.30061123  0.34490455  0.35448422]
episode: 278000, epsilon: 0.9, loss: 0.0351508, win_rate: [ 0.3007759   0.34474336  0.35448074]
episode: 280000, epsilon: 0.9, loss: 0.13776, win_rate: [ 0.30092035  0.3447202   0.35435945]
episode: 282000, epsilon: 0.9, loss: 0.0271912, win_rate: [ 0.30106985  0.34472573  0.35420442]
episode: 284000, epsilon: 0.9, loss: 0.0923988, win_rate: [ 0.30113978  0.34470301  0.3541572 ]
episode: 286000, epsilon: 0.9, loss: 0.0309716, win_rate: [ 0.30132412  0.34461768  0.3540582 ]
episode: 288000, epsilon: 0.9, loss: 0.0927393, win_rate: [ 0.3014434   0.34452311  0.35403349]
episode: 290000, epsilon: 0.9, loss: 0.0178667, win_rate: [ 0.30149551  0.34442985  0.35407464]
episode: 292000, epsilon: 0.9, loss: 0.037057, win_rate: [ 0.30160856  0.34430704  0.3540844 ]
episode: 294000, epsilon: 0.9, loss: 0.0819265, win_rate: [ 0.30177108  0.34410427  0.35412465]
episode: 296000, epsilon: 0.9, loss: 0.0592715, win_rate: [ 0.30190776  0.34418803  0.35390421]
episode: 298000, epsilon: 0.9, loss: 0.0504315, win_rate: [ 0.30205603  0.34414985  0.35379411]
episode: 300000, epsilon: 0.9, loss: 0.0286516, win_rate: [ 0.30223899  0.34405219  0.35370882]
episode: 302000, epsilon: 0.9, loss: 0.204087, win_rate: [ 0.30239635  0.34396575  0.3536379 ]
episode: 304000, epsilon: 0.9, loss: 0.0958426, win_rate: [ 0.30261085  0.34384426  0.35354489]
episode: 306000, epsilon: 0.9, loss: 0.111883, win_rate: [ 0.30275064  0.34372763  0.35352172]
episode: 308000, epsilon: 0.9, loss: 0.199278, win_rate: [ 0.30299252  0.34356382  0.35344366]
episode: 310000, epsilon: 0.9, loss: 0.103614, win_rate: [ 0.30329257  0.34347631  0.35323112]
episode: 312000, epsilon: 0.9, loss: 0.0809423, win_rate: [ 0.30340608  0.34341877  0.35317515]
episode: 314000, epsilon: 0.9, loss: 0.0548263, win_rate: [ 0.30353725  0.34336515  0.3530976 ]
episode: 316000, epsilon: 0.9, loss: 0.0386117, win_rate: [ 0.30374587  0.34320145  0.35305268]
episode: 318000, epsilon: 0.9, loss: 0.0389285, win_rate: [ 0.30369716  0.34331024  0.3529926 ]
episode: 320000, epsilon: 0.9, loss: 0.136306, win_rate: [ 0.30387405  0.34315518  0.35297077]
episode: 322000, epsilon: 0.9, loss: 0.0162871, win_rate: [ 0.30394316  0.3432567   0.35280015]
episode: 324000, epsilon: 0.9, loss: 0.0123312, win_rate: [ 0.30395894  0.34322734  0.35281373]
episode: 326000, epsilon: 0.9, loss: 0.0509802, win_rate: [ 0.30403894  0.34321674  0.35274432]
episode: 328000, epsilon: 0.9, loss: 0.129181, win_rate: [ 0.30406615  0.3431392   0.35279466]
episode: 330000, epsilon: 0.9, loss: 0.0551915, win_rate: [ 0.30410211  0.34310199  0.3527959 ]
episode: 332000, epsilon: 0.9, loss: 0.0443393, win_rate: [ 0.30416475  0.34306523  0.35277002]
episode: 334000, epsilon: 0.9, loss: 0.136224, win_rate: [ 0.30417873  0.34301095  0.35281032]
episode: 336000, epsilon: 0.9, loss: 0.182517, win_rate: [ 0.30419255  0.34288588  0.35292157]
episode: 338000, epsilon: 0.9, loss: 0.0529755, win_rate: [ 0.30422987  0.34278597  0.35298416]
episode: 340000, epsilon: 0.9, loss: 0.0434194, win_rate: [ 0.30429911  0.34268135  0.35301955]
episode: 342000, epsilon: 0.9, loss: 0.0438239, win_rate: [ 0.30433829  0.34256917  0.35309253]
episode: 344000, epsilon: 0.9, loss: 0.0497061, win_rate: [ 0.30438284  0.34246993  0.35314723]
episode: 346000, epsilon: 0.9, loss: 0.207201, win_rate: [ 0.30436328  0.34241809  0.35321863]
episode: 348000, epsilon: 0.9, loss: 0.0810478, win_rate: [ 0.30446177  0.34228637  0.35325186]
episode: 350000, epsilon: 0.9, loss: 0.0232767, win_rate: [ 0.30446199  0.34224188  0.35329613]
episode: 352000, epsilon: 0.9, loss: 0.0669126, win_rate: [ 0.3044622   0.34229448  0.35324331]
episode: 354000, epsilon: 0.9, loss: 0.0989563, win_rate: [ 0.30453304  0.34231824  0.35314872]
episode: 356000, epsilon: 0.9, loss: 0.0757773, win_rate: [ 0.30465926  0.34224342  0.35309732]
episode: 358000, epsilon: 0.9, loss: 0.0277897, win_rate: [ 0.30474217  0.34224765  0.35301019]
episode: 360000, epsilon: 0.9, loss: 0.0965361, win_rate: [ 0.30481304  0.34222405  0.35296291]
episode: 362000, epsilon: 0.9, loss: 0.10665, win_rate: [ 0.30482789  0.34225596  0.35291615]
episode: 364000, epsilon: 0.9, loss: 0.0777033, win_rate: [ 0.3048618   0.34227653  0.35286167]
episode: 366000, epsilon: 0.9, loss: 0.0140081, win_rate: [ 0.30489261  0.34222857  0.35287882]
episode: 368000, epsilon: 0.9, loss: 0.0504723, win_rate: [ 0.30496656  0.34220831  0.35282513]
episode: 370000, epsilon: 0.9, loss: 0.111004, win_rate: [ 0.30503701  0.34218016  0.35278283]
episode: 372000, epsilon: 0.9, loss: 0.0991707, win_rate: [ 0.30524649  0.34213618  0.35261733]
episode: 374000, epsilon: 0.9, loss: 0.0450218, win_rate: [ 0.30537886  0.34205256  0.35256858]
episode: 376000, epsilon: 0.9, loss: 0.0190403, win_rate: [ 0.30541674  0.34210015  0.35248311]
episode: 378000, epsilon: 0.9, loss: 0.0242864, win_rate: [ 0.30541189  0.34207846  0.35250965]
episode: 380000, epsilon: 0.9, loss: 0.047657, win_rate: [ 0.3054492   0.34202542  0.35252539]
episode: 382000, epsilon: 0.9, loss: 0.0681747, win_rate: [ 0.30546517  0.34203052  0.35250431]
episode: 384000, epsilon: 0.9, loss: 0.0903952, win_rate: [ 0.30540285  0.34209025  0.35250689]
episode: 386000, epsilon: 0.9, loss: 0.0787342, win_rate: [ 0.30543185  0.34211051  0.35245764]
episode: 388000, epsilon: 0.9, loss: 0.17759, win_rate: [ 0.30555849  0.34207644  0.35236507]
episode: 390000, epsilon: 0.9, loss: 0.0811618, win_rate: [ 0.30561973  0.34199399  0.35238628]
episode: 392000, epsilon: 0.9, loss: 0.0246235, win_rate: [ 0.30562167  0.34198382  0.35239451]
episode: 394000, epsilon: 0.9, loss: 0.0724168, win_rate: [ 0.3057378   0.34196614  0.35229606]
episode: 396000, epsilon: 0.9, loss: 0.0261151, win_rate: [ 0.30576943  0.34202186  0.35220871]
episode: 398000, epsilon: 0.9, loss: 0.0419618, win_rate: [ 0.3057128   0.34215994  0.35212726]
episode: 400000, epsilon: 0.9, loss: 0.0389928, win_rate: [ 0.30571674  0.34216664  0.35211662]
episode: 402000, epsilon: 0.9, loss: 0.117443, win_rate: [ 0.30574302  0.34208621  0.35217077]
episode: 404000, epsilon: 0.9, loss: 0.0311898, win_rate: [ 0.30576162  0.34205113  0.35218725]
episode: 406000, epsilon: 0.9, loss: 0.0517739, win_rate: [ 0.30573324  0.34200162  0.35226514]
episode: 408000, epsilon: 0.9, loss: 0.0382845, win_rate: [ 0.3058767   0.34189867  0.35222463]
episode: 410000, epsilon: 0.9, loss: 0.0194194, win_rate: [ 0.3058773   0.34190648  0.35221621]
episode: 412000, epsilon: 0.9, loss: 0.0781807, win_rate: [ 0.30596042  0.34191907  0.3521205 ]
episode: 414000, epsilon: 0.9, loss: 0.0268315, win_rate: [ 0.306021   0.3419412  0.3520378]
episode: 416000, epsilon: 0.9, loss: 0.044673, win_rate: [ 0.3060834   0.34188139  0.35203521]
episode: 418000, epsilon: 0.9, loss: 0.331834, win_rate: [ 0.30618826  0.34174799  0.35206375]
episode: 420000, epsilon: 0.9, loss: 0.0536296, win_rate: [ 0.3061826  0.3417349  0.3520825]
episode: 422000, epsilon: 0.9, loss: 0.0378078, win_rate: [ 0.30621965  0.34171483  0.35206552]
episode: 424000, epsilon: 0.9, loss: 0.0435669, win_rate: [ 0.30626579  0.34173976  0.35199445]
episode: 426000, epsilon: 0.9, loss: 0.0972483, win_rate: [ 0.30624576  0.34165647  0.35209777]
episode: 428000, epsilon: 0.9, loss: 0.0452781, win_rate: [ 0.30627265  0.34173285  0.3519945 ]
episode: 430000, epsilon: 0.9, loss: 0.017068, win_rate: [ 0.30629696  0.34177827  0.35192476]
episode: 432000, epsilon: 0.9, loss: 0.0352527, win_rate: [ 0.30636735  0.3416543   0.35197835]
episode: 434000, epsilon: 0.9, loss: 0.0870793, win_rate: [ 0.30641634  0.34169737  0.35188629]
episode: 436000, epsilon: 0.9, loss: 0.0403177, win_rate: [ 0.30641443  0.34178591  0.35179965]
episode: 438000, epsilon: 0.9, loss: 0.0226224, win_rate: [ 0.3064719   0.34176863  0.35175947]
episode: 440000, epsilon: 0.9, loss: 0.0669311, win_rate: [ 0.30658339  0.3417265   0.35169011]
episode: 442000, epsilon: 0.9, loss: 0.0893306, win_rate: [ 0.30664637  0.34168927  0.35166436]
episode: 444000, epsilon: 0.9, loss: 0.0626216, win_rate: [ 0.306693    0.34163887  0.35166813]
episode: 446000, epsilon: 0.9, loss: 0.0357122, win_rate: [ 0.30677061  0.34159789  0.3516315 ]
episode: 448000, epsilon: 0.9, loss: 0.0583008, win_rate: [ 0.30680289  0.34154834  0.35164877]
episode: 450000, epsilon: 0.9, loss: 0.0850396, win_rate: [ 0.30681265  0.34147924  0.35170811]
episode: 452000, epsilon: 0.9, loss: 0.0890502, win_rate: [ 0.30686215  0.34144393  0.35169391]
episode: 454000, epsilon: 0.9, loss: 0.0435484, win_rate: [ 0.3068958  0.3413759  0.3517283]
episode: 456000, epsilon: 0.9, loss: 0.0614234, win_rate: [ 0.30695108  0.34133916  0.35170976]
episode: 458000, epsilon: 0.9, loss: 0.212143, win_rate: [ 0.30690326  0.34135297  0.35174377]
episode: 460000, epsilon: 0.9, loss: 0.0372072, win_rate: [ 0.30693846  0.34141448  0.35164706]
episode: 462000, epsilon: 0.9, loss: 0.050082, win_rate: [ 0.30698202  0.34139104  0.35162694]
episode: 464000, epsilon: 0.9, loss: 0.0700941, win_rate: [ 0.30709417  0.34131176  0.35159407]
episode: 466000, epsilon: 0.9, loss: 0.0825135, win_rate: [ 0.3071238   0.34135334  0.35152285]
episode: 468000, epsilon: 0.9, loss: 0.0523231, win_rate: [ 0.30710404  0.34135824  0.35153771]
episode: 470000, epsilon: 0.9, loss: 0.0856748, win_rate: [ 0.3071419   0.34133757  0.35152053]
episode: 472000, epsilon: 0.9, loss: 0.0650932, win_rate: [ 0.30721333  0.34127258  0.35151409]
episode: 474000, epsilon: 0.9, loss: 0.0386773, win_rate: [ 0.30730315  0.34120181  0.35149504]
episode: 476000, epsilon: 0.9, loss: 0.0475568, win_rate: [ 0.30729767  0.34118626  0.35151607]
episode: 478000, epsilon: 0.9, loss: 0.022826, win_rate: [ 0.30734873  0.34120431  0.35144696]
episode: 480000, epsilon: 0.9, loss: 0.0322246, win_rate: [ 0.30746186  0.34111804  0.3514201 ]
episode: 482000, epsilon: 0.9, loss: 0.0829151, win_rate: [ 0.30750351  0.34106776  0.35142873]
episode: 484000, epsilon: 0.9, loss: 0.073053, win_rate: [ 0.30759441  0.34104888  0.35135671]
episode: 486000, epsilon: 0.9, loss: 0.0508463, win_rate: [ 0.30760842  0.34107955  0.35131203]
episode: 488000, epsilon: 0.9, loss: 0.0432987, win_rate: [ 0.30775347  0.34100955  0.35123699]
episode: 490000, epsilon: 0.9, loss: 0.044695, win_rate: [ 0.30788917  0.34091155  0.35119928]
episode: 492000, epsilon: 0.9, loss: 0.0176088, win_rate: [ 0.3079445   0.34089362  0.35116189]
episode: 494000, epsilon: 0.9, loss: 0.0621942, win_rate: [ 0.30806415  0.34079081  0.35114504]
episode: 496000, epsilon: 0.9, loss: 0.0405262, win_rate: [ 0.30815865  0.34078964  0.35105171]
episode: 498000, epsilon: 0.9, loss: 0.0446638, win_rate: [ 0.30824637  0.34073626  0.35101737]
episode: 500000, epsilon: 0.9, loss: 0.0184623, win_rate: [ 0.30824138  0.34071332  0.3510453 ]
episode: 502000, epsilon: 0.9, loss: 0.0599133, win_rate: [ 0.30829421  0.34070848  0.35099731]
episode: 504000, epsilon: 0.9, loss: 0.0533356, win_rate: [ 0.30833272  0.34069972  0.35096756]
episode: 506000, epsilon: 0.9, loss: 0.0336033, win_rate: [ 0.30833338  0.34072265  0.35094397]
episode: 508000, epsilon: 0.9, loss: 0.171335, win_rate: [ 0.30835569  0.34068634  0.35095797]
episode: 510000, epsilon: 0.9, loss: 0.0260027, win_rate: [ 0.30837391  0.340666    0.3509601 ]
episode: 512000, epsilon: 0.9, loss: 0.0518393, win_rate: [ 0.30842127  0.34062043  0.3509583 ]
episode: 514000, epsilon: 0.9, loss: 0.0760449, win_rate: [ 0.30844882  0.34063747  0.35091371]
episode: 516000, epsilon: 0.9, loss: 0.0549758, win_rate: [ 0.30850522  0.34064081  0.35085397]
episode: 518000, epsilon: 0.9, loss: 0.0341754, win_rate: [ 0.3085689   0.34063255  0.35079855]
episode: 520000, epsilon: 0.9, loss: 0.0795886, win_rate: [ 0.30860325  0.34058781  0.35080894]
episode: 522000, epsilon: 0.9, loss: 0.0172003, win_rate: [ 0.30863159  0.34055107  0.35081734]
episode: 524000, epsilon: 0.9, loss: 0.0352191, win_rate: [ 0.30868071  0.34044592  0.35087338]
episode: 526000, epsilon: 0.9, loss: 0.0611481, win_rate: [ 0.30876937  0.34038528  0.35084534]
episode: 528000, epsilon: 0.9, loss: 0.0598626, win_rate: [ 0.30879487  0.34032133  0.35088381]
episode: 530000, epsilon: 0.9, loss: 0.0652519, win_rate: [ 0.30882772  0.34020502  0.35096726]
episode: 532000, epsilon: 0.9, loss: 0.0702173, win_rate: [ 0.30886784  0.34019485  0.35093731]
episode: 534000, epsilon: 0.9, loss: 0.131664, win_rate: [ 0.30892639  0.3401679   0.35090571]
episode: 536000, epsilon: 0.9, loss: 0.0701256, win_rate: [ 0.3090386   0.34009638  0.35086502]
episode: 538000, epsilon: 0.9, loss: 0.101609, win_rate: [ 0.30914627  0.34006628  0.35078745]
episode: 540000, epsilon: 0.9, loss: 0.0726768, win_rate: [ 0.30916239  0.34001604  0.35082157]
episode: 542000, epsilon: 0.9, loss: 0.0563153, win_rate: [ 0.30922452  0.33998461  0.35079087]
episode: 544000, epsilon: 0.9, loss: 0.107646, win_rate: [ 0.30926781  0.33992952  0.35080266]
episode: 546000, epsilon: 0.9, loss: 0.0245504, win_rate: [ 0.30928515  0.33990048  0.35081438]
episode: 548000, epsilon: 0.9, loss: 0.0412734, win_rate: [ 0.3093644   0.33981872  0.35081688]
episode: 550000, epsilon: 0.9, loss: 0.0432466, win_rate: [ 0.30946126  0.33976665  0.35077209]
episode: 552000, epsilon: 0.9, loss: 0.109766, win_rate: [ 0.30948857  0.33973308  0.35077835]
episode: 554000, epsilon: 0.9, loss: 0.0873546, win_rate: [ 0.30958247  0.33969614  0.35072139]
episode: 556000, epsilon: 0.9, loss: 0.0537797, win_rate: [ 0.30962714  0.33966666  0.3507062 ]
episode: 558000, epsilon: 0.9, loss: 0.0662519, win_rate: [ 0.30963744  0.33970907  0.35065349]
episode: 560000, epsilon: 0.9, loss: 0.0794032, win_rate: [ 0.30971909  0.33970475  0.35057616]
episode: 562000, epsilon: 0.9, loss: 0.0910351, win_rate: [ 0.30969696  0.33970936  0.35059368]
episode: 564000, epsilon: 0.9, loss: 0.190382, win_rate: [ 0.30976541  0.33969621  0.35053839]
episode: 566000, epsilon: 0.9, loss: 0.244887, win_rate: [ 0.30982277  0.33966724  0.35050998]
episode: 568000, epsilon: 0.9, loss: 0.128635, win_rate: [ 0.3098815   0.33968602  0.35043248]
episode: 570000, epsilon: 0.9, loss: 0.0368302, win_rate: [ 0.31002402  0.33962923  0.35034675]
episode: 572000, epsilon: 0.9, loss: 0.0742465, win_rate: [ 0.31009561  0.33958157  0.35032281]
episode: 574000, epsilon: 0.9, loss: 0.0892474, win_rate: [ 0.31009702  0.33954993  0.35035305]
episode: 576000, epsilon: 0.9, loss: 0.0298162, win_rate: [ 0.31023731  0.33950462  0.35025807]
episode: 578000, epsilon: 0.9, loss: 0.105125, win_rate: [ 0.31035586  0.33950633  0.3501378 ]
episode: 580000, epsilon: 0.9, loss: 0.0965056, win_rate: [ 0.31039084  0.33953045  0.35007871]
episode: 582000, epsilon: 0.9, loss: 0.0747379, win_rate: [ 0.31050462  0.33947536  0.35002002]
episode: 584000, epsilon: 0.9, loss: 0.0460892, win_rate: [ 0.3106005   0.33942408  0.34997543]
episode: 586000, epsilon: 0.9, loss: 0.0311771, win_rate: [ 0.3106923   0.33931171  0.34999599]
episode: 588000, epsilon: 0.9, loss: 0.0223139, win_rate: [ 0.31070015  0.33929534  0.35000451]
episode: 590000, epsilon: 0.9, loss: 0.014087, win_rate: [ 0.31079947  0.33918417  0.35001636]
episode: 592000, epsilon: 0.9, loss: 0.187469, win_rate: [ 0.31084914  0.33913963  0.35001123]
episode: 594000, epsilon: 0.9, loss: 0.364125, win_rate: [ 0.31090857  0.33907014  0.3500213 ]
episode: 596000, epsilon: 0.9, loss: 0.0955397, win_rate: [ 0.3109223   0.33905144  0.35002626]
episode: 598000, epsilon: 0.9, loss: 0.0208642, win_rate: [ 0.31095935  0.33902117  0.35001948]
episode: 600000, epsilon: 0.9, loss: 0.029309, win_rate: [ 0.31099448  0.3389611   0.35004442]
episode: 602000, epsilon: 0.9, loss: 0.099173, win_rate: [ 0.31103104  0.3389496   0.35001935]
episode: 604000, epsilon: 0.9, loss: 0.0447113, win_rate: [ 0.31105412  0.3388703   0.35007558]
episode: 606000, epsilon: 0.9, loss: 0.0925096, win_rate: [ 0.31107209  0.33888888  0.35003903]
episode: 608000, epsilon: 0.9, loss: 0.0316893, win_rate: [ 0.31108501  0.33884813  0.35006686]
episode: 610000, epsilon: 0.9, loss: 0.0273971, win_rate: [ 0.31109457  0.33885026  0.35005516]
episode: 612000, epsilon: 0.9, loss: 0.0550182, win_rate: [ 0.31120537  0.33882134  0.34997328]
episode: 614000, epsilon: 0.9, loss: 0.0381274, win_rate: [ 0.31122099  0.33878446  0.34999454]
episode: 616000, epsilon: 0.9, loss: 0.0663957, win_rate: [ 0.31123488  0.33879003  0.34997508]
episode: 618000, epsilon: 0.9, loss: 0.0196721, win_rate: [ 0.31125192  0.3387535   0.34999458]
episode: 620000, epsilon: 0.9, loss: 0.0293376, win_rate: [ 0.31129305  0.33872042  0.34998653]
episode: 622000, epsilon: 0.9, loss: 0.0543059, win_rate: [ 0.31133873  0.3386972   0.34996407]
episode: 624000, epsilon: 0.9, loss: 0.0673458, win_rate: [ 0.31130078  0.33868375  0.35001546]
episode: 626000, epsilon: 0.9, loss: 0.160115, win_rate: [ 0.31139088  0.33864802  0.3499611 ]
episode: 628000, epsilon: 0.9, loss: 0.165393, win_rate: [ 0.31151224  0.33858545  0.34990231]
episode: 630000, epsilon: 0.9, loss: 0.0453944, win_rate: [ 0.31154871  0.33855026  0.34990103]
episode: 632000, epsilon: 0.9, loss: 0.106184, win_rate: [ 0.31163558  0.33848681  0.34987761]
episode: 634000, epsilon: 0.9, loss: 0.0961473, win_rate: [ 0.31165881  0.33846477  0.34987642]
episode: 636000, epsilon: 0.9, loss: 0.0491502, win_rate: [ 0.31168662  0.33844286  0.34987052]
episode: 638000, epsilon: 0.9, loss: 0.102906, win_rate: [ 0.31173305  0.33841796  0.34984898]
episode: 640000, epsilon: 0.9, loss: 0.0613453, win_rate: [ 0.31170576  0.33842916  0.34986508]
episode: 642000, epsilon: 0.9, loss: 0.0369197, win_rate: [ 0.31170356  0.33839044  0.349906  ]
episode: 644000, epsilon: 0.9, loss: 0.0469973, win_rate: [ 0.31176504  0.33835041  0.34988455]
episode: 646000, epsilon: 0.9, loss: 0.0396432, win_rate: [ 0.31178125  0.33834313  0.34987562]
episode: 648000, epsilon: 0.9, loss: 0.0310869, win_rate: [ 0.31180044  0.3383683   0.34983125]
episode: 650000, epsilon: 0.9, loss: 0.0464561, win_rate: [ 0.3117949   0.33838102  0.34982408]
episode: 652000, epsilon: 0.9, loss: 0.0306221, win_rate: [ 0.31177099  0.33840592  0.34982308]
episode: 654000, epsilon: 0.9, loss: 0.13435, win_rate: [ 0.31176558  0.33843526  0.34979916]
episode: 656000, epsilon: 0.9, loss: 0.0832305, win_rate: [ 0.31176324  0.3384507   0.34978605]
episode: 658000, epsilon: 0.9, loss: 0.0866934, win_rate: [ 0.3117974   0.33844629  0.34975631]
episode: 660000, epsilon: 0.9, loss: 0.0252544, win_rate: [ 0.31187225  0.3384404   0.34968735]
episode: 662000, epsilon: 0.9, loss: 0.0666468, win_rate: [ 0.3118953   0.33839073  0.34971397]
episode: 664000, epsilon: 0.9, loss: 0.0194304, win_rate: [ 0.31192724  0.33834136  0.3497314 ]
episode: 666000, epsilon: 0.9, loss: 0.0971439, win_rate: [ 0.31194848  0.33833132  0.3497202 ]
episode: 668000, epsilon: 0.9, loss: 0.0482307, win_rate: [ 0.31195163  0.33833033  0.34971804]
episode: 670000, epsilon: 0.9, loss: 0.0418547, win_rate: [ 0.31195177  0.33834576  0.34970246]
episode: 672000, epsilon: 0.9, loss: 0.0740064, win_rate: [ 0.3119787   0.33833283  0.34968847]
episode: 674000, epsilon: 0.9, loss: 0.122351, win_rate: [ 0.31202476  0.3382814   0.34969384]
episode: 676000, epsilon: 0.9, loss: 0.0503296, win_rate: [ 0.31208682  0.33825985  0.34965333]
episode: 678000, epsilon: 0.9, loss: 0.0287418, win_rate: [ 0.31217801  0.33823549  0.34958651]
episode: 680000, epsilon: 0.9, loss: 0.0530789, win_rate: [ 0.31224807  0.33817303  0.3495789 ]
episode: 682000, epsilon: 0.9, loss: 0.0920546, win_rate: [ 0.31226054  0.33814173  0.34959773]
episode: 684000, epsilon: 0.9, loss: 0.118793, win_rate: [ 0.31229194  0.33812524  0.34958282]
episode: 686000, epsilon: 0.9, loss: 0.0354437, win_rate: [ 0.31238293  0.33805344  0.34956363]
episode: 688000, epsilon: 0.9, loss: 0.0437258, win_rate: [ 0.31239053  0.33802422  0.34958525]
episode: 690000, epsilon: 0.9, loss: 0.0571523, win_rate: [ 0.31244592  0.33797052  0.34958355]
episode: 692000, epsilon: 0.9, loss: 0.10405, win_rate: [ 0.31248076  0.33793593  0.34958331]
episode: 694000, epsilon: 0.9, loss: 0.0227007, win_rate: [ 0.3125082   0.33789865  0.34959316]
episode: 696000, epsilon: 0.9, loss: 0.0691725, win_rate: [ 0.31252685  0.3378975   0.34957565]
episode: 698000, epsilon: 0.9, loss: 0.0441649, win_rate: [ 0.31256116  0.33786055  0.3495783 ]
episode: 700000, epsilon: 0.9, loss: 0.0424862, win_rate: [ 0.31263241  0.33782952  0.34953807]
episode: 702000, epsilon: 0.9, loss: 0.110694, win_rate: [ 0.31274172  0.3377716   0.34948668]
episode: 704000, epsilon: 0.9, loss: 0.0643624, win_rate: [ 0.31285183  0.33773674  0.34941144]
episode: 706000, epsilon: 0.9, loss: 0.019661, win_rate: [ 0.31283667  0.33770207  0.34946126]
episode: 708000, epsilon: 0.9, loss: 0.0570467, win_rate: [ 0.31283995  0.33769303  0.34946702]
episode: 710000, epsilon: 0.9, loss: 0.0218826, win_rate: [ 0.31289111  0.33764882  0.34946007]
episode: 712000, epsilon: 0.9, loss: 0.111307, win_rate: [ 0.31298271  0.33762733  0.34938996]
episode: 714000, epsilon: 0.9, loss: 0.086058, win_rate: [ 0.31300796  0.33759477  0.34939727]
episode: 716000, epsilon: 0.9, loss: 0.0836945, win_rate: [ 0.31307359  0.33755819  0.34936823]
episode: 718000, epsilon: 0.9, loss: 0.0423315, win_rate: [ 0.31311656  0.33752181  0.34936163]
episode: 720000, epsilon: 0.9, loss: 0.0469504, win_rate: [ 0.31310095  0.33753842  0.34936063]
episode: 722000, epsilon: 0.9, loss: 0.153578, win_rate: [ 0.3132004   0.33750784  0.34929176]
episode: 724000, epsilon: 0.9, loss: 0.0501421, win_rate: [ 0.31328548  0.33744429  0.34927024]
episode: 726000, epsilon: 0.9, loss: 0.127317, win_rate: [ 0.31335632  0.33738659  0.34925709]
episode: 728000, epsilon: 0.9, loss: 0.0807194, win_rate: [ 0.31342402  0.3373402   0.34923578]
episode: 730000, epsilon: 0.9, loss: 0.0983167, win_rate: [ 0.31344067  0.33729817  0.34926117]
episode: 732000, epsilon: 0.9, loss: 0.097298, win_rate: [ 0.31349411  0.33734653  0.34915936]
episode: 734000, epsilon: 0.9, loss: 0.0408951, win_rate: [ 0.31354317  0.33730063  0.3491562 ]
episode: 736000, epsilon: 0.9, loss: 0.0891946, win_rate: [ 0.31358653  0.33730253  0.34911094]
episode: 738000, epsilon: 0.9, loss: 0.164258, win_rate: [ 0.31369605  0.33721228  0.34909167]
episode: 740000, epsilon: 0.9, loss: 0.158064, win_rate: [ 0.31377795  0.33716981  0.34905223]
episode: 742000, epsilon: 0.9, loss: 0.106334, win_rate: [ 0.31382303  0.33713027  0.3490467 ]
episode: 744000, epsilon: 0.9, loss: 0.10506, win_rate: [ 0.31389474  0.33709229  0.34901297]
episode: 746000, epsilon: 0.9, loss: 0.0523631, win_rate: [ 0.31395802  0.33708668  0.3489553 ]
episode: 748000, epsilon: 0.9, loss: 0.0371306, win_rate: [ 0.31399824  0.337045    0.34895675]
episode: 750000, epsilon: 0.9, loss: 0.0303204, win_rate: [ 0.31407025  0.33700222  0.34892753]
episode: 752000, epsilon: 0.9, loss: 0.050371, win_rate: [ 0.31407139  0.33699955  0.34892906]
episode: 754000, epsilon: 0.9, loss: 0.126404, win_rate: [ 0.31408844  0.33696772  0.34894383]
episode: 756000, epsilon: 0.9, loss: 0.108825, win_rate: [ 0.31420197  0.33687654  0.3489215 ]
episode: 758000, epsilon: 0.9, loss: 0.0923645, win_rate: [ 0.31425948  0.33684784  0.34889268]
episode: 760000, epsilon: 0.9, loss: 0.115459, win_rate: [ 0.31428248  0.33680087  0.34891665]
episode: 762000, epsilon: 0.9, loss: 0.104361, win_rate: [ 0.31434211  0.33674103  0.34891686]
episode: 764000, epsilon: 0.9, loss: 0.0571901, win_rate: [ 0.3144276   0.33673123  0.34884117]
episode: 766000, epsilon: 0.9, loss: 0.103351, win_rate: [ 0.31449959  0.33668755  0.34881286]
episode: 768000, epsilon: 0.9, loss: 0.105587, win_rate: [ 0.31458032  0.33660633  0.34881335]
episode: 770000, epsilon: 0.9, loss: 0.0284438, win_rate: [ 0.31461388  0.33659826  0.34878786]
episode: 772000, epsilon: 0.9, loss: 0.181051, win_rate: [ 0.31465374  0.33659412  0.34875214]
episode: 774000, epsilon: 0.9, loss: 0.0679293, win_rate: [ 0.31466626  0.33654608  0.34878766]
episode: 776000, epsilon: 0.9, loss: 0.0728733, win_rate: [ 0.31470965  0.33648153  0.34880883]
episode: 778000, epsilon: 0.9, loss: 0.0315993, win_rate: [ 0.31478623  0.33644044  0.34877333]
episode: 780000, epsilon: 0.9, loss: 0.145858, win_rate: [ 0.31484319  0.33640854  0.34874827]
episode: 782000, epsilon: 0.9, loss: 0.0476312, win_rate: [ 0.31483591  0.3364011   0.34876298]
episode: 784000, epsilon: 0.9, loss: 0.075858, win_rate: [ 0.31486057  0.33638605  0.34875338]
episode: 786000, epsilon: 0.9, loss: 0.0693947, win_rate: [ 0.31497288  0.33633036  0.34869676]
episode: 788000, epsilon: 0.9, loss: 0.127469, win_rate: [ 0.31496534  0.33637901  0.34865565]
episode: 790000, epsilon: 0.9, loss: 0.146511, win_rate: [ 0.31502618  0.33630717  0.34866665]
episode: 792000, epsilon: 0.9, loss: 0.0327443, win_rate: [ 0.31509051  0.33629882  0.34861067]
episode: 794000, epsilon: 0.9, loss: 0.122354, win_rate: [ 0.31514066  0.33632451  0.34853483]
episode: 796000, epsilon: 0.9, loss: 0.183059, win_rate: [ 0.31520438  0.33628727  0.34850836]
episode: 798000, epsilon: 0.9, loss: 0.0890774, win_rate: [ 0.31524522  0.33621386  0.34854092]
episode: 800000, epsilon: 0.9, loss: 0.119864, win_rate: [ 0.31524836  0.33622083  0.34853081]
episode: 802000, epsilon: 0.9, loss: 0.204856, win_rate: [ 0.31534749  0.33620781  0.3484447 ]
episode: 804000, epsilon: 0.9, loss: 0.114617, win_rate: [ 0.31540011  0.33617122  0.34842867]
episode: 806000, epsilon: 0.9, loss: 0.0747776, win_rate: [ 0.31542889  0.33612117  0.34844994]
episode: 808000, epsilon: 0.9, loss: 0.0746621, win_rate: [ 0.31547238  0.33606023  0.34846739]
episode: 810000, epsilon: 0.9, loss: 0.0772573, win_rate: [ 0.31551443  0.3360107   0.34847488]
episode: 812000, epsilon: 0.9, loss: 0.0218551, win_rate: [ 0.31558089  0.33598973  0.34842937]
episode: 814000, epsilon: 0.9, loss: 0.079478, win_rate: [ 0.3156286   0.3359615   0.34840989]
episode: 816000, epsilon: 0.9, loss: 0.132875, win_rate: [ 0.31565893  0.33595547  0.3483856 ]
episode: 818000, epsilon: 0.9, loss: 0.0548598, win_rate: [ 0.31567688  0.33594458  0.34837855]
episode: 820000, epsilon: 0.9, loss: 0.0506202, win_rate: [ 0.31569352  0.33592276  0.34838372]
episode: 822000, epsilon: 0.9, loss: 0.0910465, win_rate: [ 0.31576361  0.3358974   0.34833899]
episode: 824000, epsilon: 0.9, loss: 0.197816, win_rate: [ 0.31577146  0.3358928   0.34833574]
episode: 826000, epsilon: 0.9, loss: 0.15523, win_rate: [ 0.31577323  0.33592211  0.34830466]
episode: 828000, epsilon: 0.9, loss: 0.293722, win_rate: [ 0.31587039  0.33593438  0.34819523]
episode: 830000, epsilon: 0.9, loss: 0.32723, win_rate: [ 0.3159442  0.335902   0.3481538]
episode: 832000, epsilon: 0.9, loss: 0.0613427, win_rate: [ 0.31599481  0.33590464  0.34810054]
episode: 834000, epsilon: 0.9, loss: 0.0843161, win_rate: [ 0.31610154  0.3358653   0.34803316]
episode: 836000, epsilon: 0.9, loss: 0.0840188, win_rate: [ 0.31618981  0.33583333  0.34797686]
episode: 838000, epsilon: 0.9, loss: 0.223547, win_rate: [ 0.31626215  0.33579435  0.3479435 ]
episode: 840000, epsilon: 0.9, loss: 0.391604, win_rate: [ 0.31638296  0.33577103  0.34784601]
episode: 842000, epsilon: 0.9, loss: 0.149702, win_rate: [ 0.3165115   0.33573594  0.34775256]
episode: 844000, epsilon: 0.9, loss: 0.196039, win_rate: [ 0.31663825  0.33568207  0.34767968]
episode: 846000, epsilon: 0.9, loss: 0.678151, win_rate: [ 0.31674312  0.33564499  0.34761188]
episode: 848000, epsilon: 0.9, loss: 0.130465, win_rate: [ 0.3168534   0.33560338  0.34754322]
episode: 850000, epsilon: 0.9, loss: 0.530904, win_rate: [ 0.31698433  0.33555725  0.34745841]
episode: 852000, epsilon: 0.9, loss: 1.05845, win_rate: [ 0.31717686  0.33548317  0.34733997]
episode: 854000, epsilon: 0.9, loss: 0.294223, win_rate: [ 0.31731462  0.33541647  0.34726891]
episode: 856000, epsilon: 0.9, loss: 0.24601, win_rate: [ 0.31748211  0.33535475  0.34716315]
episode: 858000, epsilon: 0.9, loss: 0.140483, win_rate: [ 0.31763949  0.33532129  0.34703922]
episode: 860000, epsilon: 0.9, loss: 0.259827, win_rate: [ 0.31778451  0.33527868  0.34693681]
episode: 862000, epsilon: 0.9, loss: 0.569678, win_rate: [ 0.31796715  0.33519335  0.3468395 ]
episode: 864000, epsilon: 0.9, loss: 0.173579, win_rate: [ 0.31808528  0.33516281  0.34675191]
episode: 866000, epsilon: 0.9, loss: 0.122627, win_rate: [ 0.31815899  0.33514395  0.34669706]
episode: 868000, epsilon: 0.9, loss: 0.109343, win_rate: [ 0.31826922  0.33509754  0.34663324]
episode: 870000, epsilon: 0.9, loss: 0.063344, win_rate: [ 0.31835711  0.33501801  0.34662489]
episode: 872000, epsilon: 0.9, loss: 0.100988, win_rate: [ 0.31836775  0.33501796  0.34661428]
episode: 874000, epsilon: 0.9, loss: 0.0945025, win_rate: [ 0.31843785  0.33498474  0.34657741]
episode: 876000, epsilon: 0.9, loss: 0.0959731, win_rate: [ 0.31846197  0.33497108  0.34656696]
episode: 878000, epsilon: 0.9, loss: 0.0977433, win_rate: [ 0.31854406  0.3349495   0.34650644]
episode: 880000, epsilon: 0.9, loss: 0.0577249, win_rate: [ 0.31858487  0.33488598  0.34652915]
episode: 882000, epsilon: 0.9, loss: 0.0436054, win_rate: [ 0.31865157  0.33484429  0.34650414]
episode: 884000, epsilon: 0.9, loss: 0.180819, win_rate: [ 0.31869308  0.33483333  0.34647359]
episode: 886000, epsilon: 0.9, loss: 0.141343, win_rate: [ 0.31870619  0.33485402  0.34643979]
episode: 888000, epsilon: 0.9, loss: 0.0563448, win_rate: [ 0.31871698  0.33490503  0.34637799]
episode: 890000, epsilon: 0.9, loss: 0.0563546, win_rate: [ 0.31870751  0.33489625  0.34639624]
episode: 892000, epsilon: 0.9, loss: 0.080206, win_rate: [ 0.31877879  0.33487182  0.34634939]
episode: 894000, epsilon: 0.9, loss: 0.098336, win_rate: [ 0.31881061  0.33486763  0.34632176]
episode: 896000, epsilon: 0.9, loss: 0.103839, win_rate: [ 0.31886348  0.33483445  0.34630207]
episode: 898000, epsilon: 0.9, loss: 0.0273521, win_rate: [ 0.31889942  0.33479807  0.34630251]
episode: 900000, epsilon: 0.9, loss: 0.177287, win_rate: [ 0.3189752   0.33475629  0.3462685 ]
episode: 902000, epsilon: 0.9, loss: 0.106376, win_rate: [ 0.31904954  0.33473466  0.3462158 ]
episode: 904000, epsilon: 0.9, loss: 0.0525703, win_rate: [ 0.31914235  0.33474299  0.34611466]
episode: 906000, epsilon: 0.9, loss: 0.0541536, win_rate: [ 0.31919832  0.33471817  0.3460835 ]
episode: 908000, epsilon: 0.9, loss: 0.111993, win_rate: [ 0.31922212  0.33466593  0.34611195]
episode: 910000, epsilon: 0.9, loss: 0.115006, win_rate: [ 0.31923921  0.33463699  0.3461238 ]
episode: 912000, epsilon: 0.9, loss: 0.299983, win_rate: [ 0.31927926  0.33462463  0.34609611]
episode: 914000, epsilon: 0.9, loss: 0.0933144, win_rate: [ 0.31934648  0.3345981   0.34605542]
episode: 916000, epsilon: 0.9, loss: 0.0335358, win_rate: [ 0.31943742  0.33453784  0.34602473]
episode: 918000, epsilon: 0.9, loss: 0.0534133, win_rate: [ 0.31950292  0.33451924  0.34597784]
episode: 920000, epsilon: 0.9, loss: 0.298455, win_rate: [ 0.319604    0.33445181  0.34594419]
episode: 922000, epsilon: 0.9, loss: 0.0622094, win_rate: [ 0.31963197  0.33441287  0.34595516]
episode: 924000, epsilon: 0.9, loss: 0.0879177, win_rate: [ 0.31967173  0.33438925  0.34593902]
episode: 926000, epsilon: 0.9, loss: 0.0941045, win_rate: [ 0.31970376  0.33437005  0.34592619]
episode: 928000, epsilon: 0.9, loss: 0.0475058, win_rate: [ 0.31972703  0.3343671   0.34590588]
episode: 930000, epsilon: 0.9, loss: 0.0578048, win_rate: [ 0.31977385  0.3343448   0.34588135]
episode: 932000, epsilon: 0.9, loss: 0.0390147, win_rate: [ 0.31981082  0.33431616  0.34587302]
episode: 934000, epsilon: 0.9, loss: 0.0726372, win_rate: [ 0.3197866   0.33431228  0.34590113]
episode: 936000, epsilon: 0.9, loss: 0.111728, win_rate: [ 0.31982551  0.33427956  0.34589493]
episode: 938000, epsilon: 0.9, loss: 0.0625941, win_rate: [ 0.31986     0.33427043  0.34586957]
episode: 940000, epsilon: 0.9, loss: 0.0299137, win_rate: [ 0.31994753  0.33425496  0.3457975 ]
episode: 942000, epsilon: 0.9, loss: 0.0726181, win_rate: [ 0.3199986   0.33424911  0.34575229]
episode: 944000, epsilon: 0.9, loss: 0.038456, win_rate: [ 0.32001661  0.33422104  0.34576235]
episode: 946000, epsilon: 0.9, loss: 0.0437996, win_rate: [ 0.32001235  0.33424912  0.34573854]
episode: 948000, epsilon: 0.9, loss: 0.0963162, win_rate: [ 0.32005135  0.33425492  0.34569373]
episode: 950000, epsilon: 0.9, loss: 0.14459, win_rate: [ 0.32011861  0.33420596  0.34567543]
episode: 952000, epsilon: 0.9, loss: 0.0497625, win_rate: [ 0.32014462  0.33417402  0.34568136]
episode: 954000, epsilon: 0.9, loss: 0.167269, win_rate: [ 0.32016528  0.33414954  0.34568517]
episode: 956000, epsilon: 0.9, loss: 0.105935, win_rate: [ 0.32019841  0.33410007  0.34570152]
episode: 958000, epsilon: 0.9, loss: 0.0334828, win_rate: [ 0.32023662  0.33407481  0.34568857]
episode: 960000, epsilon: 0.9, loss: 0.023265, win_rate: [ 0.32026946  0.33406319  0.34566735]
episode: 962000, epsilon: 0.9, loss: 0.131661, win_rate: [ 0.32030632  0.33405059  0.34564309]
episode: 964000, epsilon: 0.9, loss: 0.0965433, win_rate: [ 0.32033473  0.33401833  0.34564694]
episode: 966000, epsilon: 0.9, loss: 0.101683, win_rate: [ 0.32037027  0.33399137  0.34563836]
episode: 968000, epsilon: 0.9, loss: 0.118052, win_rate: [ 0.32042942  0.33396866  0.34560192]
episode: 970000, epsilon: 0.9, loss: 0.0608048, win_rate: [ 0.32048936  0.33392852  0.34558212]
episode: 972000, epsilon: 0.9, loss: 0.177394, win_rate: [ 0.32050584  0.33390809  0.34558606]
episode: 974000, epsilon: 0.9, loss: 0.135373, win_rate: [ 0.32056641  0.3338744   0.34555919]
episode: 976000, epsilon: 0.9, loss: 0.0473096, win_rate: [ 0.32062262  0.33382548  0.3455519 ]
episode: 978000, epsilon: 0.9, loss: 0.0492055, win_rate: [ 0.32066532  0.33381459  0.3455201 ]
episode: 980000, epsilon: 0.9, loss: 0.0584045, win_rate: [ 0.32068641  0.33378537  0.34552822]
episode: 982000, epsilon: 0.9, loss: 0.0643864, win_rate: [ 0.3206789   0.33379905  0.34552205]
episode: 984000, epsilon: 0.9, loss: 0.0388458, win_rate: [ 0.32072224  0.33376287  0.34551489]
episode: 986000, epsilon: 0.9, loss: 0.0367076, win_rate: [ 0.32071874  0.33379784  0.34548342]
episode: 988000, epsilon: 0.9, loss: 0.0652551, win_rate: [ 0.32074158  0.33379015  0.34546827]
episode: 990000, epsilon: 0.9, loss: 0.0554728, win_rate: [ 0.32077139  0.33376936  0.34545925]
episode: 992000, epsilon: 0.9, loss: 0.124443, win_rate: [ 0.32079504  0.33376277  0.34544219]
episode: 994000, epsilon: 0.9, loss: 0.0114861, win_rate: [ 0.32081758  0.33374514  0.34543728]
episode: 996000, epsilon: 0.9, loss: 0.0524915, win_rate: [ 0.32084305  0.3337075   0.34544945]
episode: 998000, epsilon: 0.9, loss: 0.0256955, win_rate: [ 0.32087543  0.33369506  0.34542951]
episode: 1000000, epsilon: 0.9, loss: 0.0506189, win_rate: [ 0.32090168  0.33365167  0.34544665]
episode: 1002000, epsilon: 0.9, loss: 0.0637978, win_rate: [ 0.3209378   0.33363639  0.3454258 ]
episode: 1004000, epsilon: 0.9, loss: 0.133739, win_rate: [ 0.3209688   0.33362118  0.34541001]
episode: 1006000, epsilon: 0.9, loss: 0.0498048, win_rate: [ 0.32095992  0.33364679  0.34539329]
episode: 1008000, epsilon: 0.9, loss: 0.0782023, win_rate: [ 0.32100464  0.33361673  0.34537863]
episode: 1010000, epsilon: 0.9, loss: 0.0723313, win_rate: [ 0.32100859  0.3336066   0.34538481]
episode: 1012000, epsilon: 0.9, loss: 0.0862523, win_rate: [ 0.32100364  0.33364097  0.34535539]
episode: 1014000, epsilon: 0.9, loss: 0.0841308, win_rate: [ 0.32105984  0.33363774  0.34530242]
episode: 1016000, epsilon: 0.9, loss: 0.053612, win_rate: [ 0.32109319  0.33360991  0.3452969 ]
episode: 1018000, epsilon: 0.9, loss: 0.086102, win_rate: [ 0.32114409  0.33359889  0.34525703]
episode: 1020000, epsilon: 0.9, loss: 0.186626, win_rate: [ 0.32116733  0.33358595  0.34524672]
episode: 1022000, epsilon: 0.9, loss: 0.109487, win_rate: [ 0.32122865  0.33355643  0.34521493]
episode: 1024000, epsilon: 0.9, loss: 0.0703795, win_rate: [ 0.3212282   0.33352799  0.3452438 ]
episode: 1026000, epsilon: 0.9, loss: 0.197689, win_rate: [ 0.32124628  0.33353379  0.34521994]
episode: 1028000, epsilon: 0.9, loss: 0.0935955, win_rate: [ 0.32132848  0.333488    0.34518352]
episode: 1030000, epsilon: 0.9, loss: 0.125539, win_rate: [ 0.32138124  0.33347346  0.3451453 ]
episode: 1032000, epsilon: 0.9, loss: 0.0677419, win_rate: [ 0.32145124  0.33343572  0.34511304]
episode: 1034000, epsilon: 0.9, loss: 0.139671, win_rate: [ 0.32149099  0.33340974  0.34509928]
episode: 1036000, epsilon: 0.9, loss: 0.275071, win_rate: [ 0.32154892  0.33340701  0.34504407]
episode: 1038000, epsilon: 0.9, loss: 0.0970106, win_rate: [ 0.32165191  0.33337058  0.34497751]
episode: 1040000, epsilon: 0.9, loss: 0.198754, win_rate: [ 0.32170161  0.33334776  0.34495063]
episode: 1042000, epsilon: 0.9, loss: 0.124288, win_rate: [ 0.32175017  0.33330774  0.34494209]
episode: 1044000, epsilon: 0.9, loss: 0.128581, win_rate: [ 0.32182632  0.33327746  0.34489622]
episode: 1046000, epsilon: 0.9, loss: 0.0742144, win_rate: [ 0.32189071  0.33326259  0.34484671]
episode: 1048000, epsilon: 0.9, loss: 0.111558, win_rate: [ 0.32190236  0.33326399  0.34483364]
episode: 1050000, epsilon: 0.9, loss: 0.0842839, win_rate: [ 0.32191398  0.33327968  0.34480634]
episode: 1052000, epsilon: 0.9, loss: 0.0381279, win_rate: [ 0.32191794  0.33328961  0.34479245]
episode: 1054000, epsilon: 0.9, loss: 0.0547885, win_rate: [ 0.32196649  0.33327293  0.34476058]
episode: 1056000, epsilon: 0.9, loss: 0.0934835, win_rate: [ 0.32198738  0.3332762   0.34473642]
episode: 1058000, epsilon: 0.9, loss: 0.0308796, win_rate: [ 0.32200442  0.33327095  0.34472463]
episode: 1060000, epsilon: 0.9, loss: 0.0676955, win_rate: [ 0.32207045  0.33325063  0.34467892]
episode: 1062000, epsilon: 0.9, loss: 0.0999288, win_rate: [ 0.32209103  0.33324545  0.34466352]
episode: 1064000, epsilon: 0.9, loss: 0.0832386, win_rate: [ 0.32210214  0.33320927  0.34468859]
episode: 1066000, epsilon: 0.9, loss: 0.0532691, win_rate: [ 0.3221573   0.33319293  0.34464977]
episode: 1068000, epsilon: 0.9, loss: 0.0614725, win_rate: [ 0.32223284  0.33316823  0.34459893]
episode: 1070000, epsilon: 0.9, loss: 0.0927298, win_rate: [ 0.32226045  0.33315203  0.34458753]
episode: 1072000, epsilon: 0.9, loss: 0.0552366, win_rate: [ 0.32229821  0.33315454  0.34454725]
episode: 1074000, epsilon: 0.9, loss: 0.0715846, win_rate: [ 0.32230882  0.33316636  0.34452482]
episode: 1076000, epsilon: 0.9, loss: 0.124786, win_rate: [ 0.32234821  0.33317255  0.34447923]
episode: 1078000, epsilon: 0.9, loss: 0.0572503, win_rate: [ 0.32236612  0.33319542  0.34443846]
episode: 1080000, epsilon: 0.9, loss: 0.0553883, win_rate: [ 0.32240248  0.33317469  0.34442283]
episode: 1082000, epsilon: 0.9, loss: 0.485112, win_rate: [ 0.32246088  0.33314941  0.3443897 ]
episode: 1084000, epsilon: 0.9, loss: 0.062188, win_rate: [ 0.32251723  0.33312331  0.34435946]
episode: 1086000, epsilon: 0.9, loss: 0.148255, win_rate: [ 0.32255587  0.33314242  0.34430171]
episode: 1088000, epsilon: 0.9, loss: 0.0851735, win_rate: [ 0.32265687  0.33309252  0.3442506 ]
episode: 1090000, epsilon: 0.9, loss: 0.080351, win_rate: [ 0.32273732  0.33306483  0.34419785]
episode: 1092000, epsilon: 0.9, loss: 0.0814694, win_rate: [ 0.32285135  0.33301618  0.34413247]
episode: 1094000, epsilon: 0.9, loss: 0.100962, win_rate: [ 0.32291104  0.33297959  0.34410937]
episode: 1096000, epsilon: 0.9, loss: 0.207704, win_rate: [ 0.32300609  0.33291028  0.34408363]
episode: 1098000, epsilon: 0.9, loss: 0.142077, win_rate: [ 0.32308076  0.33289041  0.34402883]
episode: 1100000, epsilon: 0.9, loss: 0.0911651, win_rate: [ 0.32314971  0.33284242  0.34400787]
episode: 1102000, epsilon: 0.9, loss: 0.128663, win_rate: [ 0.32322384  0.33281095  0.34396521]
episode: 1104000, epsilon: 0.9, loss: 0.0518107, win_rate: [ 0.32327145  0.33277234  0.34395621]
episode: 1106000, epsilon: 0.9, loss: 0.218895, win_rate: [ 0.3232809   0.33274744  0.34397166]
episode: 1108000, epsilon: 0.9, loss: 0.0427323, win_rate: [ 0.32333545  0.3326766   0.34398796]
episode: 1110000, epsilon: 0.9, loss: 0.0568478, win_rate: [ 0.3233862   0.33264925  0.34396455]
episode: 1112000, epsilon: 0.9, loss: 0.130261, win_rate: [ 0.32344216  0.33259772  0.34396012]
episode: 1114000, epsilon: 0.9, loss: 0.14201, win_rate: [ 0.3234638   0.33259485  0.34394134]
episode: 1116000, epsilon: 0.9, loss: 0.119496, win_rate: [ 0.32352749  0.33254271  0.3439298 ]
episode: 1118000, epsilon: 0.9, loss: 0.0743658, win_rate: [ 0.32355874  0.33250239  0.34393887]
episode: 1120000, epsilon: 0.9, loss: 0.0912209, win_rate: [ 0.323589    0.33248095  0.34393005]
episode: 1122000, epsilon: 0.9, loss: 0.111783, win_rate: [ 0.32361647  0.33247118  0.34391235]
episode: 1124000, epsilon: 0.9, loss: 0.09344, win_rate: [ 0.32363405  0.33247301  0.34389293]
episode: 1126000, epsilon: 0.9, loss: 0.0460583, win_rate: [ 0.32366845  0.33243043  0.34390112]
episode: 1128000, epsilon: 0.9, loss: 0.304033, win_rate: [ 0.32370805  0.33241815  0.34387381]
episode: 1130000, epsilon: 0.9, loss: 0.119355, win_rate: [ 0.32377228  0.33241564  0.34381209]
episode: 1132000, epsilon: 0.9, loss: 0.36513, win_rate: [ 0.32380978  0.33238575  0.34380447]
episode: 1134000, epsilon: 0.9, loss: 0.528388, win_rate: [ 0.32387626  0.3323692   0.34375455]
episode: 1136000, epsilon: 0.9, loss: 0.0765519, win_rate: [ 0.3239557   0.33235094  0.34369336]
episode: 1138000, epsilon: 0.9, loss: 0.234509, win_rate: [ 0.32399093  0.33236702  0.34364205]
episode: 1140000, epsilon: 0.9, loss: 0.0926188, win_rate: [ 0.32406901  0.33231638  0.34361461]
episode: 1142000, epsilon: 0.9, loss: 0.121914, win_rate: [ 0.32413982  0.3323202   0.34353998]
episode: 1144000, epsilon: 0.9, loss: 0.281578, win_rate: [ 0.32419465  0.3323109   0.34349445]
episode: 1146000, epsilon: 0.9, loss: 0.34272, win_rate: [ 0.32425277  0.33231734  0.34342989]
episode: 1148000, epsilon: 0.9, loss: 0.119814, win_rate: [ 0.32432463  0.33230198  0.34337339]
episode: 1150000, epsilon: 0.9, loss: 0.44163, win_rate: [ 0.32439972  0.33223449  0.34336579]
episode: 1152000, epsilon: 0.9, loss: 0.137115, win_rate: [ 0.32444503  0.3322202   0.34333477]
episode: 1154000, epsilon: 0.9, loss: 0.192059, win_rate: [ 0.32451532  0.33219729  0.3432874 ]
episode: 1156000, epsilon: 0.9, loss: 0.274369, win_rate: [ 0.32461823  0.33215283  0.34322894]
episode: 1158000, epsilon: 0.9, loss: 0.345975, win_rate: [ 0.32472424  0.33209816  0.3431776 ]
episode: 1160000, epsilon: 0.9, loss: 0.224558, win_rate: [ 0.3247842   0.33206523  0.34315057]
episode: 1162000, epsilon: 0.9, loss: 0.314646, win_rate: [ 0.3248276   0.33205909  0.3431133 ]
episode: 1164000, epsilon: 0.9, loss: 0.248279, win_rate: [ 0.3248889   0.33201346  0.34309764]
episode: 1166000, epsilon: 0.9, loss: 0.112562, win_rate: [ 0.32495427  0.33199714  0.34304859]
episode: 1168000, epsilon: 0.9, loss: 0.573, win_rate: [ 0.32501513  0.3319552   0.34302967]
episode: 1170000, epsilon: 0.9, loss: 0.176453, win_rate: [ 0.32504502  0.33191937  0.3430356 ]
episode: 1172000, epsilon: 0.9, loss: 0.0706293, win_rate: [ 0.32510638  0.33190501  0.34298862]
episode: 1174000, epsilon: 0.9, loss: 0.165791, win_rate: [ 0.32516838  0.33188302  0.3429486 ]
episode: 1176000, epsilon: 0.9, loss: 0.277321, win_rate: [ 0.32521656  0.33187727  0.34290617]
episode: 1178000, epsilon: 0.9, loss: 0.176294, win_rate: [ 0.32529259  0.33186305  0.34284436]
episode: 1180000, epsilon: 0.9, loss: 0.189686, win_rate: [ 0.32538532  0.33184294  0.34277174]
episode: 1182000, epsilon: 0.9, loss: 0.104302, win_rate: [ 0.32543712  0.33180683  0.34275606]
episode: 1184000, epsilon: 0.9, loss: 0.0868109, win_rate: [ 0.32547608  0.33176999  0.34275393]
episode: 1186000, epsilon: 0.9, loss: 0.0652112, win_rate: [ 0.3255402   0.33174846  0.34271135]
episode: 1188000, epsilon: 0.9, loss: 0.135256, win_rate: [ 0.32558306  0.33171437  0.34270257]
episode: 1190000, epsilon: 0.9, loss: 0.0837329, win_rate: [ 0.32564847  0.33166947  0.34268206]
episode: 1192000, epsilon: 0.9, loss: 0.227607, win_rate: [ 0.32567674  0.33164318  0.34268008]
episode: 1194000, epsilon: 0.9, loss: 0.0755312, win_rate: [ 0.32570073  0.33162368  0.34267559]
episode: 1196000, epsilon: 0.9, loss: 0.0790938, win_rate: [ 0.32573886  0.33157832  0.34268282]
episode: 1198000, epsilon: 0.9, loss: 0.0662919, win_rate: [ 0.3258211   0.33153729  0.34264162]
episode: 1200000, epsilon: 0.9, loss: 0.163289, win_rate: [ 0.32587723  0.33148306  0.34263971]
episode: 1202000, epsilon: 0.9, loss: 0.0807239, win_rate: [ 0.32588991  0.33148725  0.34262284]
episode: 1204000, epsilon: 0.9, loss: 0.0745174, win_rate: [ 0.32590421  0.3314964   0.34259938]
episode: 1206000, epsilon: 0.9, loss: 0.112053, win_rate: [ 0.32589774  0.33148812  0.34261414]
episode: 1208000, epsilon: 0.9, loss: 0.139617, win_rate: [ 0.3259095   0.33148069  0.34260982]
episode: 1210000, epsilon: 0.9, loss: 0.0172084, win_rate: [ 0.32593609  0.33147576  0.34258815]
episode: 1212000, epsilon: 0.9, loss: 0.0569144, win_rate: [ 0.32593537  0.33146837  0.34259625]
episode: 1214000, epsilon: 0.9, loss: 0.0942227, win_rate: [ 0.32594866  0.33147749  0.34257385]
episode: 1216000, epsilon: 0.9, loss: 0.0475396, win_rate: [ 0.32597424  0.33143476  0.342591  ]
episode: 1218000, epsilon: 0.9, loss: 0.079552, win_rate: [ 0.32600137  0.33143076  0.34256786]
episode: 1220000, epsilon: 0.9, loss: 0.143652, win_rate: [ 0.32602514  0.33141284  0.34256201]
episode: 1222000, epsilon: 0.9, loss: 0.0665064, win_rate: [ 0.32604229  0.33140889  0.34254882]
episode: 1224000, epsilon: 0.9, loss: 0.057487, win_rate: [ 0.32606754  0.3313543   0.34257815]
episode: 1226000, epsilon: 0.9, loss: 0.0724877, win_rate: [ 0.32607641  0.33133986  0.34258373]
episode: 1228000, epsilon: 0.9, loss: 0.121565, win_rate: [ 0.32608035  0.33132302  0.34259663]
episode: 1230000, epsilon: 0.9, loss: 0.174685, win_rate: [ 0.32608347  0.33132168  0.34259484]
episode: 1232000, epsilon: 0.9, loss: 0.107148, win_rate: [ 0.32611987  0.33129356  0.34258657]
episode: 1234000, epsilon: 0.9, loss: 0.200395, win_rate: [ 0.32615533  0.3312704   0.34257428]
episode: 1236000, epsilon: 0.9, loss: 0.389671, win_rate: [ 0.32617368  0.33123679  0.34258953]
episode: 1238000, epsilon: 0.9, loss: 0.0708498, win_rate: [ 0.3261936   0.33123398  0.34257242]
episode: 1240000, epsilon: 0.9, loss: 0.111774, win_rate: [ 0.32622635  0.33121828  0.34255537]
episode: 1242000, epsilon: 0.9, loss: 0.117341, win_rate: [ 0.32625175  0.33118411  0.34256414]
episode: 1244000, epsilon: 0.9, loss: 0.176517, win_rate: [ 0.32631405  0.33116694  0.34251902]
episode: 1246000, epsilon: 0.9, loss: 0.0704494, win_rate: [ 0.32635207  0.3311466   0.34250133]
episode: 1248000, epsilon: 0.9, loss: 0.106966, win_rate: [ 0.32633788  0.33115759  0.34250453]
episode: 1250000, epsilon: 0.9, loss: 0.176652, win_rate: [ 0.32636214  0.33115254  0.34248533]
episode: 1252000, epsilon: 0.9, loss: 0.0328003, win_rate: [ 0.32638712  0.33112194  0.34249094]
episode: 1254000, epsilon: 0.9, loss: 0.0466095, win_rate: [ 0.32644791  0.33108427  0.34246783]
episode: 1256000, epsilon: 0.9, loss: 0.0586598, win_rate: [ 0.3265085   0.33107219  0.34241931]
episode: 1258000, epsilon: 0.9, loss: 0.0606739, win_rate: [ 0.32656969  0.33106412  0.34236618]
episode: 1260000, epsilon: 0.9, loss: 0.0840457, win_rate: [ 0.32658228  0.33104021  0.34237751]
episode: 1262000, epsilon: 0.9, loss: 0.171472, win_rate: [ 0.32660196  0.33102272  0.34237532]
episode: 1264000, epsilon: 0.9, loss: 0.0829915, win_rate: [ 0.3266374   0.33101635  0.34234625]
episode: 1266000, epsilon: 0.9, loss: 0.0695489, win_rate: [ 0.32669011  0.33098236  0.34232753]
episode: 1268000, epsilon: 0.9, loss: 0.161013, win_rate: [ 0.32670085  0.3309816   0.34231755]
episode: 1270000, epsilon: 0.9, loss: 0.118508, win_rate: [ 0.3267336   0.33097297  0.34229343]
episode: 1272000, epsilon: 0.9, loss: 0.154057, win_rate: [ 0.32674424  0.33097222  0.34228354]
episode: 1274000, epsilon: 0.9, loss: 0.0649897, win_rate: [ 0.3268145  0.330955   0.3422305]
episode: 1276000, epsilon: 0.9, loss: 0.10683, win_rate: [ 0.32684771  0.33093704  0.34221525]
episode: 1278000, epsilon: 0.9, loss: 0.0455086, win_rate: [ 0.32689255  0.33091445  0.342193  ]
episode: 1280000, epsilon: 0.9, loss: 0.151251, win_rate: [ 0.32695834  0.33086849  0.34217317]
episode: 1282000, epsilon: 0.9, loss: 0.084946, win_rate: [ 0.3269857   0.33088118  0.34213312]
episode: 1284000, epsilon: 0.9, loss: 0.0567544, win_rate: [ 0.32700753  0.33088058  0.34211188]
episode: 1286000, epsilon: 0.9, loss: 0.150667, win_rate: [ 0.3270464   0.33086055  0.34209305]
episode: 1288000, epsilon: 0.9, loss: 0.0427269, win_rate: [ 0.32709835  0.33083437  0.34206728]
episode: 1290000, epsilon: 0.9, loss: 0.052844, win_rate: [ 0.32714006  0.33082455  0.34203539]
episode: 1292000, epsilon: 0.9, loss: 0.0999404, win_rate: [ 0.32715222  0.33081631  0.34203147]
episode: 1294000, epsilon: 0.9, loss: 0.100458, win_rate: [ 0.32714967  0.33085446  0.34199587]
episode: 1296000, epsilon: 0.9, loss: 0.0162908, win_rate: [ 0.32716487  0.33084851  0.34198662]
episode: 1298000, epsilon: 0.9, loss: 0.0293832, win_rate: [ 0.32720622  0.33085028  0.3419435 ]
episode: 1300000, epsilon: 0.9, loss: 0.0712604, win_rate: [ 0.32723667  0.33084667  0.34191666]
episode: 1302000, epsilon: 0.9, loss: 0.0269455, win_rate: [ 0.32723631  0.33083231  0.34193138]
episode: 1304000, epsilon: 0.9, loss: 0.0648866, win_rate: [ 0.32727352  0.33081646  0.34191001]
episode: 1306000, epsilon: 0.9, loss: 0.12663, win_rate: [ 0.32731139  0.33079454  0.34189407]
episode: 1308000, epsilon: 0.9, loss: 0.0854803, win_rate: [ 0.32735067  0.33077115  0.34187818]
episode: 1310000, epsilon: 0.9, loss: 0.0435073, win_rate: [ 0.32739212  0.33076311  0.34184478]
episode: 1312000, epsilon: 0.9, loss: 0.0570287, win_rate: [ 0.32741515  0.33074746  0.34183739]
episode: 1314000, epsilon: 0.9, loss: 0.0342732, win_rate: [ 0.32743506  0.33073567  0.34182927]
episode: 1316000, epsilon: 0.9, loss: 0.0467514, win_rate: [ 0.32744504  0.33071783  0.34183713]
episode: 1318000, epsilon: 0.9, loss: 0.0426302, win_rate: [ 0.3274694   0.33072054  0.34181006]
episode: 1320000, epsilon: 0.9, loss: 0.0279334, win_rate: [ 0.32746263  0.33071717  0.3418202 ]
episode: 1322000, epsilon: 0.9, loss: 0.0408862, win_rate: [ 0.32748084  0.33071155  0.34180761]
episode: 1324000, epsilon: 0.9, loss: 0.100278, win_rate: [ 0.32752317  0.33070292  0.34177391]
episode: 1326000, epsilon: 0.9, loss: 0.136517, win_rate: [ 0.3275533   0.33066566  0.34178104]
episode: 1328000, epsilon: 0.9, loss: 0.0484754, win_rate: [ 0.32761045  0.3306368   0.34175275]
episode: 1330000, epsilon: 0.9, loss: 0.0814314, win_rate: [ 0.32762457  0.3306193   0.34175613]
episode: 1332000, epsilon: 0.9, loss: 0.277086, win_rate: [ 0.32762813  0.33063489  0.34173698]
episode: 1334000, epsilon: 0.9, loss: 0.0892482, win_rate: [ 0.32767367  0.33061894  0.34170739]
episode: 1336000, epsilon: 0.9, loss: 0.0639606, win_rate: [ 0.3277026   0.33059332  0.34170409]
episode: 1338000, epsilon: 0.9, loss: 0.0646907, win_rate: [ 0.32772845  0.33057225  0.3416993 ]
episode: 1340000, epsilon: 0.9, loss: 0.0388254, win_rate: [ 0.32776319  0.33057289  0.34166392]
episode: 1342000, epsilon: 0.9, loss: 0.0636873, win_rate: [ 0.32778515  0.33054446  0.34167039]
episode: 1344000, epsilon: 0.9, loss: 0.0374163, win_rate: [ 0.32776389  0.33054216  0.34169394]
episode: 1346000, epsilon: 0.9, loss: 0.076964, win_rate: [ 0.32778356  0.33054062  0.34167582]
episode: 1348000, epsilon: 0.9, loss: 0.0546578, win_rate: [ 0.32776459  0.33054501  0.3416904 ]
episode: 1350000, epsilon: 0.9, loss: 0.132656, win_rate: [ 0.32776272  0.33057013  0.34166715]
episode: 1352000, epsilon: 0.9, loss: 0.0389217, win_rate: [ 0.32782002  0.33054931  0.34163066]
episode: 1354000, epsilon: 0.9, loss: 0.105668, win_rate: [ 0.32782694  0.33053373  0.34163933]
episode: 1356000, epsilon: 0.9, loss: 0.0957844, win_rate: [ 0.32783899  0.33053884  0.34162217]
episode: 1358000, epsilon: 0.9, loss: 0.0583298, win_rate: [ 0.32788636  0.33051596  0.34159769]
episode: 1360000, epsilon: 0.9, loss: 0.0839057, win_rate: [ 0.32791226  0.3305049   0.34158284]
episode: 1362000, epsilon: 0.9, loss: 0.063126, win_rate: [ 0.32793001  0.33049388  0.34157611]
episode: 1364000, epsilon: 0.9, loss: 0.151012, win_rate: [ 0.32796237  0.33048143  0.3415562 ]
episode: 1366000, epsilon: 0.9, loss: 0.1048, win_rate: [ 0.32797926  0.33047926  0.34154148]
episode: 1368000, epsilon: 0.9, loss: 0.143023, win_rate: [ 0.32801365  0.33045005  0.3415363 ]
episode: 1370000, epsilon: 0.9, loss: 0.0938775, win_rate: [ 0.32806107  0.3304472   0.34149172]
episode: 1372000, epsilon: 0.9, loss: 0.0361067, win_rate: [ 0.32806755  0.33043781  0.34149465]
episode: 1374000, epsilon: 0.9, loss: 0.1212, win_rate: [ 0.32807618  0.33044736  0.34147646]
episode: 1376000, epsilon: 0.9, loss: 0.16099, win_rate: [ 0.32810369  0.33041982  0.3414765 ]
episode: 1378000, epsilon: 0.9, loss: 0.0868403, win_rate: [ 0.32812966  0.33043009  0.34144025]
episode: 1380000, epsilon: 0.9, loss: 0.0384476, win_rate: [ 0.32812585  0.33042585  0.3414483 ]
episode: 1382000, epsilon: 0.9, loss: 0.0260155, win_rate: [ 0.32813001  0.3304303   0.3414397 ]
episode: 1384000, epsilon: 0.9, loss: 0.103647, win_rate: [ 0.32816739  0.33042317  0.34140944]
episode: 1386000, epsilon: 0.9, loss: 0.0384848, win_rate: [ 0.32820106  0.33039154  0.3414074 ]
episode: 1388000, epsilon: 0.9, loss: 0.0797451, win_rate: [ 0.32825409  0.33038161  0.34136431]
episode: 1390000, epsilon: 0.9, loss: 0.0524445, win_rate: [ 0.32826955  0.33038178  0.34134868]
episode: 1392000, epsilon: 0.9, loss: 0.105575, win_rate: [ 0.32829143  0.33035537  0.3413532 ]
episode: 1394000, epsilon: 0.9, loss: 0.0964302, win_rate: [ 0.32832258  0.33036849  0.34130894]
episode: 1396000, epsilon: 0.9, loss: 0.660192, win_rate: [ 0.32837441  0.33034933  0.34127626]
episode: 1398000, epsilon: 0.9, loss: 0.440172, win_rate: [ 0.32842108  0.33032594  0.34125297]
episode: 1400000, epsilon: 0.9, loss: 0.401245, win_rate: [ 0.32849905  0.33032405  0.3411769 ]
episode: 1402000, epsilon: 0.9, loss: 0.138815, win_rate: [ 0.32858821  0.33029292  0.34111887]
episode: 1404000, epsilon: 0.9, loss: 0.199474, win_rate: [ 0.32867569  0.33026686  0.34105745]
episode: 1406000, epsilon: 0.9, loss: 0.0616103, win_rate: [ 0.32873945  0.33026292  0.34099762]
episode: 1408000, epsilon: 0.9, loss: 0.264443, win_rate: [ 0.32879522  0.33024124  0.34096354]
episode: 1410000, epsilon: 0.9, loss: 0.220737, win_rate: [ 0.32885509  0.33022033  0.34092458]
episode: 1412000, epsilon: 0.9, loss: 0.414904, win_rate: [ 0.32892611  0.33018744  0.34088644]
episode: 1414000, epsilon: 0.9, loss: 0.133153, win_rate: [ 0.32896653  0.33016243  0.34087105]
episode: 1416000, epsilon: 0.9, loss: 0.0521375, win_rate: [ 0.32902237  0.33012406  0.34085357]
episode: 1418000, epsilon: 0.9, loss: 0.101729, win_rate: [ 0.32905195  0.3301126   0.34083544]
episode: 1420000, epsilon: 0.9, loss: 0.0727152, win_rate: [ 0.32907723  0.33011878  0.34080399]
episode: 1422000, epsilon: 0.9, loss: 0.0580982, win_rate: [ 0.32911791  0.33009892  0.34078316]
episode: 1424000, epsilon: 0.9, loss: 0.0896811, win_rate: [ 0.32909176  0.33011915  0.34078909]
episode: 1426000, epsilon: 0.9, loss: 0.0547686, win_rate: [ 0.32909935  0.33010917  0.34079149]
episode: 1428000, epsilon: 0.9, loss: 0.203545, win_rate: [ 0.32911181  0.3300873   0.34080088]
episode: 1430000, epsilon: 0.9, loss: 0.0455078, win_rate: [ 0.32914872  0.33005711  0.34079417]
episode: 1432000, epsilon: 0.9, loss: 0.0676725, win_rate: [ 0.32918622  0.33003818  0.3407756 ]
episode: 1434000, epsilon: 0.9, loss: 0.101674, win_rate: [ 0.32921386  0.33002836  0.34075778]
episode: 1436000, epsilon: 0.9, loss: 0.111394, win_rate: [ 0.32925465  0.33001509  0.34073026]
episode: 1438000, epsilon: 0.9, loss: 0.114793, win_rate: [ 0.32925707  0.33001855  0.34072438]
episode: 1440000, epsilon: 0.9, loss: 0.084687, win_rate: [ 0.32927477  0.3299956   0.34072962]
episode: 1442000, epsilon: 0.9, loss: 0.0312413, win_rate: [ 0.32928549  0.32999076  0.34072376]
episode: 1444000, epsilon: 0.9, loss: 0.0771281, win_rate: [ 0.32932595  0.32997276  0.34070129]
episode: 1446000, epsilon: 0.9, loss: 0.0480501, win_rate: [ 0.32934555  0.32996519  0.34068925]
episode: 1448000, epsilon: 0.9, loss: 0.0368343, win_rate: [ 0.32936787  0.329939    0.34069313]
episode: 1450000, epsilon: 0.9, loss: 0.0611446, win_rate: [ 0.32939908  0.32992322  0.3406777 ]
episode: 1452000, epsilon: 0.9, loss: 0.185039, win_rate: [ 0.32942746  0.32990817  0.34066437]
episode: 1454000, epsilon: 0.9, loss: 0.127926, win_rate: [ 0.32947364  0.32989936  0.340627  ]
episode: 1456000, epsilon: 0.9, loss: 0.296776, win_rate: [ 0.32949291  0.32989675  0.34061034]
episode: 1458000, epsilon: 0.9, loss: 0.23728, win_rate: [ 0.329508    0.32988592  0.34060608]
episode: 1460000, epsilon: 0.9, loss: 0.140718, win_rate: [ 0.32952238  0.32989566  0.34058196]
episode: 1462000, epsilon: 0.9, loss: 0.137101, win_rate: [ 0.32954765  0.32989581  0.34055654]
episode: 1464000, epsilon: 0.9, loss: 0.195209, win_rate: [ 0.32957901  0.32988161  0.34053938]
episode: 1466000, epsilon: 0.9, loss: 0.111786, win_rate: [ 0.32960892  0.32986403  0.34052705]
episode: 1468000, epsilon: 0.9, loss: 0.138932, win_rate: [ 0.32963057  0.32982811  0.34054132]
episode: 1470000, epsilon: 0.9, loss: 0.143663, win_rate: [ 0.32967937  0.32979773  0.3405229 ]
episode: 1472000, epsilon: 0.9, loss: 0.211779, win_rate: [ 0.32973279  0.32976676  0.34050045]
episode: 1474000, epsilon: 0.9, loss: 0.149264, win_rate: [ 0.32975419  0.3297569   0.34048891]
episode: 1476000, epsilon: 0.9, loss: 0.065541, win_rate: [ 0.3298094   0.32972403  0.34046657]
episode: 1478000, epsilon: 0.9, loss: 0.249781, win_rate: [ 0.32984281  0.32971223  0.34044497]
episode: 1480000, epsilon: 0.9, loss: 0.276795, win_rate: [ 0.3298741   0.32969707  0.34042882]
episode: 1482000, epsilon: 0.9, loss: 0.227075, win_rate: [ 0.32994512  0.32968601  0.34036887]
episode: 1484000, epsilon: 0.9, loss: 0.0795458, win_rate: [ 0.32999506  0.32964803  0.34035691]
episode: 1486000, epsilon: 0.9, loss: 0.0401089, win_rate: [ 0.33005025  0.32960543  0.34034432]
episode: 1488000, epsilon: 0.9, loss: 0.0542639, win_rate: [ 0.33009857  0.32959521  0.34030622]
episode: 1490000, epsilon: 0.9, loss: 0.141791, win_rate: [ 0.33014542  0.32958233  0.34027225]
episode: 1492000, epsilon: 0.9, loss: 0.451235, win_rate: [ 0.33016265  0.32959093  0.34024642]
episode: 1494000, epsilon: 0.9, loss: 0.0919776, win_rate: [ 0.33018987  0.32960018  0.34020995]
episode: 1496000, epsilon: 0.9, loss: 0.343012, win_rate: [ 0.33025847  0.32960072  0.34014082]
episode: 1498000, epsilon: 0.9, loss: 0.587568, win_rate: [ 0.33031153  0.32957388  0.34011459]
episode: 1500000, epsilon: 0.9, loss: 0.150008, win_rate: [ 0.33037511  0.32955045  0.34007444]
episode: 1502000, epsilon: 0.9, loss: 0.183544, win_rate: [ 0.33044785  0.32950311  0.34004904]
episode: 1504000, epsilon: 0.9, loss: 0.134447, win_rate: [ 0.33048116  0.32949513  0.34002371]
episode: 1506000, epsilon: 0.9, loss: 0.127789, win_rate: [ 0.33052302  0.32949447  0.33998251]
episode: 1508000, epsilon: 0.9, loss: 0.193506, win_rate: [ 0.3305714   0.32947127  0.33995733]
episode: 1510000, epsilon: 0.9, loss: 0.66819, win_rate: [ 0.33062892  0.32944945  0.33992163]
episode: 1512000, epsilon: 0.9, loss: 0.511293, win_rate: [ 0.33067505  0.32945018  0.33987478]
episode: 1514000, epsilon: 0.9, loss: 0.117619, win_rate: [ 0.33073558  0.3294225   0.33984192]
episode: 1516000, epsilon: 0.9, loss: 0.0548076, win_rate: [ 0.33078474  0.32941271  0.33980255]
learning_rate = 0.0001

episode: 1518000, epsilon: 0.9, loss: 0, win_rate: [ 0.39666667  0.28333333  0.32      ]
episode: 1520000, epsilon: 0.9, loss: 0.271768, win_rate: [ 0.3773913   0.31434783  0.30826087]
episode: 1522000, epsilon: 0.9, loss: 0.203776, win_rate: [ 0.3744186  0.3155814  0.31     ]
episode: 1524000, epsilon: 0.9, loss: 0.117541, win_rate: [ 0.37507937  0.3152381   0.30968254]
episode: 1526000, epsilon: 0.9, loss: 0.372409, win_rate: [ 0.37614458  0.31843373  0.30542169]
episode: 1528000, epsilon: 0.9, loss: 0.838946, win_rate: [ 0.37621359  0.31466019  0.30912621]
episode: 1530000, epsilon: 0.9, loss: 0.267074, win_rate: [ 0.37544715  0.31585366  0.30869919]
episode: 1532000, epsilon: 0.9, loss: 0.262787, win_rate: [ 0.37265734  0.31902098  0.30832168]
episode: 1534000, epsilon: 0.9, loss: 0.383413, win_rate: [ 0.37312883  0.31588957  0.3109816 ]
episode: 1536000, epsilon: 0.9, loss: 0.193851, win_rate: [ 0.37174863  0.31590164  0.31234973]
episode: 1538000, epsilon: 0.9, loss: 0.694638, win_rate: [ 0.37408867  0.31453202  0.31137931]
episode: 1540000, epsilon: 0.9, loss: 0.393239, win_rate: [ 0.37251121  0.3167713   0.31071749]
episode: 1542000, epsilon: 0.9, loss: 0.394909, win_rate: [ 0.37386831  0.31641975  0.30971193]
episode: 1544000, epsilon: 0.9, loss: 0.174821, win_rate: [ 0.37429658  0.31596958  0.30973384]
episode: 1546000, epsilon: 0.9, loss: 0.197844, win_rate: [ 0.37459364  0.31660777  0.30879859]
episode: 1548000, epsilon: 0.9, loss: 0.160928, win_rate: [ 0.37405941  0.31706271  0.30887789]
episode: 1550000, epsilon: 0.9, loss: 0.214234, win_rate: [ 0.37532508  0.31668731  0.30798762]
episode: 1552000, epsilon: 0.9, loss: 0.303717, win_rate: [ 0.37548105  0.31629738  0.30822157]
episode: 1554000, epsilon: 0.9, loss: 0.431463, win_rate: [ 0.37752066  0.31553719  0.30694215]
episode: 1556000, epsilon: 0.9, loss: 0.154005, win_rate: [ 0.37879896  0.31331593  0.30788512]
episode: 1558000, epsilon: 0.9, loss: 0.171552, win_rate: [ 0.378933    0.31344913  0.30761787]
episode: 1560000, epsilon: 0.9, loss: 0.0993965, win_rate: [ 0.37992908  0.3127896   0.30728132]
episode: 1562000, epsilon: 0.9, loss: 0.201021, win_rate: [ 0.37923251  0.3131377   0.3076298 ]
episode: 1564000, epsilon: 0.9, loss: 0.421224, win_rate: [ 0.37928726  0.3137581   0.30695464]
episode: 1566000, epsilon: 0.9, loss: 0.120805, win_rate: [ 0.37929607  0.31397516  0.30672878]
episode: 1568000, epsilon: 0.9, loss: 0.107305, win_rate: [ 0.37942346  0.31367793  0.30689861]
episode: 1570000, epsilon: 0.9, loss: 0.0587219, win_rate: [ 0.37873805  0.31418738  0.30707457]
episode: 1572000, epsilon: 0.9, loss: 0.220529, win_rate: [ 0.3779558   0.31499079  0.30705341]
episode: 1574000, epsilon: 0.9, loss: 0.242412, win_rate: [ 0.37721137  0.31516874  0.30761989]
episode: 1576000, epsilon: 0.9, loss: 0.184262, win_rate: [ 0.37636364  0.31524871  0.30838765]
episode: 1578000, epsilon: 0.9, loss: 0.0950626, win_rate: [ 0.37573798  0.31505804  0.30920398]
episode: 1580000, epsilon: 0.9, loss: 0.20343, win_rate: [ 0.37536116  0.31483146  0.30980738]
episode: 1582000, epsilon: 0.9, loss: 0.13025, win_rate: [ 0.3748056   0.31482115  0.31037325]
episode: 1584000, epsilon: 0.9, loss: 0.0490923, win_rate: [ 0.37457014  0.31485671  0.31057315]
episode: 1586000, epsilon: 0.9, loss: 0.0573395, win_rate: [ 0.37440703  0.31487555  0.31071742]
episode: 1588000, epsilon: 0.9, loss: 0.0926311, win_rate: [ 0.37401138  0.31516358  0.31082504]
episode: 1590000, epsilon: 0.9, loss: 0.162171, win_rate: [ 0.37318119  0.31536653  0.31145228]
episode: 1592000, epsilon: 0.9, loss: 0.0699946, win_rate: [ 0.37269179  0.31522207  0.31208614]
episode: 1594000, epsilon: 0.9, loss: 0.201559, win_rate: [ 0.37218873  0.31503277  0.31277851]
episode: 1596000, epsilon: 0.9, loss: 0.163621, win_rate: [ 0.37162197  0.31526181  0.31311622]
episode: 1598000, epsilon: 0.9, loss: 0.0798386, win_rate: [ 0.37085928  0.31572852  0.3134122 ]
episode: 1600000, epsilon: 0.9, loss: 0.135935, win_rate: [ 0.37076549  0.31562576  0.31360875]
episode: 1602000, epsilon: 0.9, loss: 0.0463097, win_rate: [ 0.3709134   0.31559905  0.31348754]
episode: 1604000, epsilon: 0.9, loss: 0.279938, win_rate: [ 0.37041715  0.31543453  0.31414832]
episode: 1606000, epsilon: 0.9, loss: 0.113932, win_rate: [ 0.36983012  0.31558324  0.31458664]
episode: 1608000, epsilon: 0.9, loss: 0.176885, win_rate: [ 0.36952381  0.31578073  0.31469546]
episode: 1610000, epsilon: 0.9, loss: 0.29787, win_rate: [ 0.36937161  0.31600217  0.31462622]
episode: 1612000, epsilon: 0.9, loss: 0.498863, win_rate: [ 0.36949099  0.31583245  0.31467656]
episode: 1614000, epsilon: 0.9, loss: 0.257677, win_rate: [ 0.36970924  0.31559709  0.31469367]
episode: 1616000, epsilon: 0.9, loss: 0.295472, win_rate: [ 0.36957274  0.31547304  0.31495422]
episode: 1618000, epsilon: 0.9, loss: 0.111942, win_rate: [ 0.36936191  0.31547358  0.31516451]
episode: 1620000, epsilon: 0.9, loss: 0.142979, win_rate: [ 0.36923754  0.31542522  0.31533724]
episode: 1622000, epsilon: 0.9, loss: 0.165517, win_rate: [ 0.36889741  0.31563758  0.315465  ]
episode: 1624000, epsilon: 0.9, loss: 0.134166, win_rate: [ 0.36862653  0.31552211  0.31585136]
episode: 1626000, epsilon: 0.9, loss: 0.129667, win_rate: [ 0.36855032  0.31551247  0.31593721]
episode: 1628000, epsilon: 0.9, loss: 0.114139, win_rate: [ 0.36827743  0.31567543  0.31604714]
episode: 1630000, epsilon: 0.9, loss: 0.266667, win_rate: [ 0.36785396  0.31612645  0.31601959]
episode: 1632000, epsilon: 0.9, loss: 0.0479422, win_rate: [ 0.36776903  0.31622922  0.31600175]
episode: 1634000, epsilon: 0.9, loss: 0.0508065, win_rate: [ 0.36769561  0.31651763  0.31578676]
episode: 1636000, epsilon: 0.9, loss: 0.130412, win_rate: [ 0.36797126  0.3162891   0.31573964]
episode: 1638000, epsilon: 0.9, loss: 0.202116, win_rate: [ 0.36788861  0.31621779  0.3158936 ]
episode: 1640000, epsilon: 0.9, loss: 0.0633848, win_rate: [ 0.36780049  0.31615699  0.31604252]
episode: 1642000, epsilon: 0.9, loss: 0.314197, win_rate: [ 0.36759453  0.31625101  0.31615447]
episode: 1644000, epsilon: 0.9, loss: 0.111878, win_rate: [ 0.36769596  0.31585115  0.31645289]
episode: 1646000, epsilon: 0.9, loss: 0.0471003, win_rate: [ 0.36749805  0.3158145   0.31668745]
episode: 1648000, epsilon: 0.9, loss: 0.292445, win_rate: [ 0.36710668  0.31607061  0.31682272]
episode: 1650000, epsilon: 0.9, loss: 0.0798865, win_rate: [ 0.36705215  0.31606198  0.31688587]
episode: 1652000, epsilon: 0.9, loss: 0.138114, win_rate: [ 0.36708861  0.31576322  0.31714818]
episode: 1654000, epsilon: 0.9, loss: 0.315876, win_rate: [ 0.36717535  0.31562729  0.31719736]
episode: 1656000, epsilon: 0.9, loss: 0.171169, win_rate: [ 0.36699928  0.31540853  0.31759219]
episode: 1658000, epsilon: 0.9, loss: 0.28435, win_rate: [ 0.36685674  0.31547398  0.31766928]
episode: 1660000, epsilon: 0.9, loss: 0.0685767, win_rate: [ 0.36645116  0.31550246  0.31804638]
episode: 1662000, epsilon: 0.9, loss: 0.109366, win_rate: [ 0.36604297  0.31550243  0.31845461]
episode: 1664000, epsilon: 0.9, loss: 0.131903, win_rate: [ 0.36588517  0.31558442  0.31853042]
episode: 1666000, epsilon: 0.9, loss: 0.0375194, win_rate: [ 0.36572488  0.3158058   0.31846932]
episode: 1668000, epsilon: 0.9, loss: 0.0401955, win_rate: [ 0.36566201  0.31580173  0.31853626]
episode: 1670000, epsilon: 0.9, loss: 0.105899, win_rate: [ 0.3657912   0.31575181  0.31845699]
episode: 1672000, epsilon: 0.9, loss: 0.0251195, win_rate: [ 0.36550227  0.31577447  0.31872327]
episode: 1674000, epsilon: 0.9, loss: 0.0564443, win_rate: [ 0.36548305  0.3156238   0.31889315]
episode: 1676000, epsilon: 0.9, loss: 0.0894084, win_rate: [ 0.36509792  0.3156475   0.31925458]
episode: 1678000, epsilon: 0.9, loss: 0.151455, win_rate: [ 0.36530256  0.31557704  0.3191204 ]
episode: 1680000, epsilon: 0.9, loss: 0.171882, win_rate: [ 0.36548367  0.31556377  0.31895256]
episode: 1682000, epsilon: 0.9, loss: 0.0587437, win_rate: [ 0.36537432  0.31550822  0.31911747]
episode: 1684000, epsilon: 0.9, loss: 0.196431, win_rate: [ 0.36549008  0.31546001  0.31904991]
episode: 1686000, epsilon: 0.9, loss: 0.129804, win_rate: [ 0.36551396  0.3154486   0.31903743]
episode: 1688000, epsilon: 0.9, loss: 0.0954819, win_rate: [ 0.36531415  0.31564885  0.31903699]
episode: 1690000, epsilon: 0.9, loss: 0.0985823, win_rate: [ 0.36512478  0.31583285  0.31904237]
episode: 1692000, epsilon: 0.9, loss: 0.109748, win_rate: [ 0.36508319  0.31596672  0.31895009]
episode: 1694000, epsilon: 0.9, loss: 0.164812, win_rate: [ 0.36457175  0.31633012  0.31909813]
episode: 1696000, epsilon: 0.9, loss: 0.0174163, win_rate: [ 0.36443634  0.31632081  0.31924285]
episode: 1698000, epsilon: 0.9, loss: 0.0357966, win_rate: [ 0.36426511  0.31642818  0.31930671]
episode: 1700000, epsilon: 0.9, loss: 0.122127, win_rate: [ 0.36443225  0.31629183  0.31927592]
episode: 1702000, epsilon: 0.9, loss: 0.199405, win_rate: [ 0.36436788  0.31628323  0.31934889]
episode: 1704000, epsilon: 0.9, loss: 0.0911045, win_rate: [ 0.36439077  0.31624799  0.31936125]
episode: 1706000, epsilon: 0.9, loss: 0.0538207, win_rate: [ 0.36431227  0.31631439  0.31937334]
episode: 1708000, epsilon: 0.9, loss: 0.106255, win_rate: [ 0.36423542  0.31644246  0.31932212]
episode: 1710000, epsilon: 0.9, loss: 0.0608051, win_rate: [ 0.36426937  0.31624025  0.31949038]
episode: 1712000, epsilon: 0.9, loss: 0.282458, win_rate: [ 0.36427689  0.31613999  0.31958312]
episode: 1714000, epsilon: 0.9, loss: 0.277402, win_rate: [ 0.36412634  0.31624045  0.31963321]
episode: 1716000, epsilon: 0.9, loss: 0.0979924, win_rate: [ 0.36420071  0.31609682  0.31970247]
episode: 1718000, epsilon: 0.9, loss: 0.0389021, win_rate: [ 0.36429855  0.316001    0.31970045]
episode: 1720000, epsilon: 0.9, loss: 0.109107, win_rate: [ 0.36451804  0.31593673  0.31954523]
episode: 1722000, epsilon: 0.9, loss: 0.0496547, win_rate: [ 0.36456192  0.31586882  0.31956926]
episode: 1724000, epsilon: 0.9, loss: 0.151681, win_rate: [ 0.36449346  0.31571013  0.31979641]
episode: 1726000, epsilon: 0.9, loss: 0.0802187, win_rate: [ 0.36451752  0.31567451  0.31980797]
episode: 1728000, epsilon: 0.9, loss: 0.268461, win_rate: [ 0.36440799  0.31573466  0.31985735]
episode: 1730000, epsilon: 0.9, loss: 0.0275447, win_rate: [ 0.36450306  0.31564767  0.31984927]
episode: 1732000, epsilon: 0.9, loss: 0.433142, win_rate: [ 0.36448437  0.31559963  0.31991601]
episode: 1734000, epsilon: 0.9, loss: 0.0565796, win_rate: [ 0.36442441  0.31576514  0.31981045]
episode: 1736000, epsilon: 0.9, loss: 0.102539, win_rate: [ 0.36434265  0.31573523  0.31992213]
episode: 1738000, epsilon: 0.9, loss: 0.0571739, win_rate: [ 0.36425783  0.31588743  0.31985474]
episode: 1740000, epsilon: 0.9, loss: 0.428564, win_rate: [ 0.36408007  0.31613135  0.31978857]
episode: 1742000, epsilon: 0.9, loss: 0.0735125, win_rate: [ 0.36425769  0.31592956  0.31981275]
episode: 1744000, epsilon: 0.9, loss: 0.176237, win_rate: [ 0.36422448  0.31610252  0.319673  ]
episode: 1746000, epsilon: 0.9, loss: 0.0640948, win_rate: [ 0.36420061  0.31618484  0.31961454]
episode: 1748000, epsilon: 0.9, loss: 0.0436814, win_rate: [ 0.36428571  0.31591837  0.31979592]
episode: 1750000, epsilon: 0.9, loss: 0.141155, win_rate: [ 0.36436935  0.3158545   0.31977615]
episode: 1752000, epsilon: 0.9, loss: 0.488514, win_rate: [ 0.36438754  0.31580452  0.31980794]
episode: 1754000, epsilon: 0.9, loss: 0.118051, win_rate: [ 0.36439695  0.3157427   0.31986035]
episode: 1756000, epsilon: 0.9, loss: 0.0881868, win_rate: [ 0.36444398  0.31572388  0.31983214]
episode: 1758000, epsilon: 0.9, loss: 0.0655707, win_rate: [ 0.36454016  0.31565959  0.31980025]
episode: 1760000, epsilon: 0.9, loss: 0.160123, win_rate: [ 0.36459348  0.31572431  0.31968221]
episode: 1762000, epsilon: 0.9, loss: 0.0511614, win_rate: [ 0.36468277  0.31553418  0.31978305]
episode: 1764000, epsilon: 0.9, loss: 0.102161, win_rate: [ 0.36482745  0.31549736  0.31967519]
episode: 1766000, epsilon: 0.9, loss: 0.156944, win_rate: [ 0.36475634  0.31550946  0.31973419]
episode: 1768000, epsilon: 0.9, loss: 0.0964088, win_rate: [ 0.36471434  0.3156692   0.31961646]
episode: 1770000, epsilon: 0.9, loss: 0.245925, win_rate: [ 0.36473246  0.3157828   0.31948474]
episode: 1772000, epsilon: 0.9, loss: 0.347117, win_rate: [ 0.36472277  0.31582776  0.31944947]
episode: 1774000, epsilon: 0.9, loss: 0.111248, win_rate: [ 0.36474054  0.31577448  0.31948498]
episode: 1776000, epsilon: 0.9, loss: 0.094318, win_rate: [ 0.36464576  0.31572977  0.31962447]
episode: 1778000, epsilon: 0.9, loss: 0.488849, win_rate: [ 0.36460238  0.3155705   0.31982712]
episode: 1780000, epsilon: 0.9, loss: 0.0401163, win_rate: [ 0.3647236  0.3153679  0.3199085]
episode: 1782000, epsilon: 0.9, loss: 0.161964, win_rate: [ 0.36478244  0.3153235   0.31989406]
episode: 1784000, epsilon: 0.9, loss: 0.0703916, win_rate: [ 0.36464138  0.31533609  0.32002253]
episode: 1786000, epsilon: 0.9, loss: 0.117756, win_rate: [ 0.36469624  0.31538949  0.31991428]
episode: 1788000, epsilon: 0.9, loss: 0.768996, win_rate: [ 0.36472438  0.3154458   0.31982982]
episode: 1790000, epsilon: 0.9, loss: 0.197246, win_rate: [ 0.3647264   0.31534704  0.31992655]
episode: 1792000, epsilon: 0.9, loss: 0.174123, win_rate: [ 0.36462268  0.31542836  0.31994896]
episode: 1794000, epsilon: 0.9, loss: 0.0557632, win_rate: [ 0.36467608  0.31540717  0.31991676]
episode: 1796000, epsilon: 0.9, loss: 0.069438, win_rate: [ 0.36467481  0.31531082  0.32001437]
episode: 1798000, epsilon: 0.9, loss: 0.0502161, win_rate: [ 0.36476632  0.31514449  0.32008919]
episode: 1800000, epsilon: 0.9, loss: 0.142659, win_rate: [ 0.36491321  0.31506199  0.3200248 ]
episode: 1802000, epsilon: 0.9, loss: 0.0907863, win_rate: [ 0.36465353  0.31533591  0.32001055]
episode: 1804000, epsilon: 0.9, loss: 0.0863359, win_rate: [ 0.36455816  0.31543137  0.32001048]
episode: 1806000, epsilon: 0.9, loss: 0.108273, win_rate: [ 0.36460284  0.31535553  0.32004162]
episode: 1808000, epsilon: 0.9, loss: 0.0247519, win_rate: [ 0.36444368  0.31537375  0.32018257]
episode: 1810000, epsilon: 0.9, loss: 0.0219992, win_rate: [ 0.36437906  0.3154362   0.32018474]
episode: 1812000, epsilon: 0.9, loss: 0.076335, win_rate: [ 0.36413863  0.31558953  0.32027183]
episode: 1814000, epsilon: 0.9, loss: 0.0567257, win_rate: [ 0.36403645  0.31563618  0.32032737]
episode: 1816000, epsilon: 0.9, loss: 0.236187, win_rate: [ 0.36372779  0.31570902  0.32056319]
episode: 1818000, epsilon: 0.9, loss: 0.117301, win_rate: [ 0.36368298  0.31587746  0.32043956]
episode: 1820000, epsilon: 0.9, loss: 0.105495, win_rate: [ 0.36368177  0.31588819  0.32043004]
episode: 1822000, epsilon: 0.9, loss: 0.0317167, win_rate: [ 0.36361157  0.31589878  0.32048965]
episode: 1824000, epsilon: 0.9, loss: 0.256244, win_rate: [ 0.36351616  0.31591903  0.32056481]
episode: 1826000, epsilon: 0.9, loss: 0.0709017, win_rate: [ 0.36345443  0.31593578  0.3206098 ]
episode: 1828000, epsilon: 0.9, loss: 0.0558197, win_rate: [ 0.36320657  0.31608443  0.32070899]
episode: 1830000, epsilon: 0.9, loss: 0.277857, win_rate: [ 0.3631284  0.3160935  0.3207781]
episode: 1832000, epsilon: 0.9, loss: 0.067033, win_rate: [ 0.36293032  0.31614699  0.32092269]
episode: 1834000, epsilon: 0.9, loss: 0.0423557, win_rate: [ 0.3629276   0.31609864  0.32097376]
episode: 1836000, epsilon: 0.9, loss: 0.0785001, win_rate: [ 0.36307257  0.31608231  0.32084511]
episode: 1838000, epsilon: 0.9, loss: 0.26868, win_rate: [ 0.36313768  0.31614112  0.3207212 ]
episode: 1840000, epsilon: 0.9, loss: 0.14206, win_rate: [ 0.36311511  0.31619919  0.3206857 ]
episode: 1842000, epsilon: 0.9, loss: 0.0585054, win_rate: [ 0.36292939  0.31638915  0.32068147]
episode: 1844000, epsilon: 0.9, loss: 0.282725, win_rate: [ 0.36297579  0.31633466  0.32068955]
episode: 1846000, epsilon: 0.9, loss: 0.0894541, win_rate: [ 0.36293329  0.31628693  0.32077977]
episode: 1848000, epsilon: 0.9, loss: 0.111059, win_rate: [ 0.36296397  0.31634575  0.32069028]
episode: 1850000, epsilon: 0.9, loss: 0.119615, win_rate: [ 0.36301234  0.31633163  0.32065603]
episode: 1852000, epsilon: 0.9, loss: 0.0815712, win_rate: [ 0.36298534  0.3163117   0.32070296]
episode: 1854000, epsilon: 0.9, loss: 0.032212, win_rate: [ 0.36296759  0.31641392  0.3206185 ]
episode: 1856000, epsilon: 0.9, loss: 0.0587608, win_rate: [ 0.36303281  0.31640556  0.32056163]
episode: 1858000, epsilon: 0.9, loss: 0.301586, win_rate: [ 0.36303262  0.31636497  0.32060241]
episode: 1860000, epsilon: 0.9, loss: 0.261731, win_rate: [ 0.3630967   0.31635408  0.32054923]
episode: 1862000, epsilon: 0.9, loss: 0.0796917, win_rate: [ 0.3632704   0.31624455  0.32048504]
episode: 1864000, epsilon: 0.9, loss: 0.303258, win_rate: [ 0.36330927  0.31615362  0.32053711]
episode: 1866000, epsilon: 0.9, loss: 0.0417984, win_rate: [ 0.36332759  0.31608671  0.3205857 ]
episode: 1868000, epsilon: 0.9, loss: 0.141981, win_rate: [ 0.36327434  0.31615187  0.32057379]
episode: 1870000, epsilon: 0.9, loss: 0.115921, win_rate: [ 0.3632671   0.31625319  0.3204797 ]
episode: 1872000, epsilon: 0.9, loss: 0.764199, win_rate: [ 0.36333333  0.31618402  0.32048264]
episode: 1874000, epsilon: 0.9, loss: 0.135364, win_rate: [ 0.36330059  0.31612125  0.32057816]
episode: 1876000, epsilon: 0.9, loss: 0.0612705, win_rate: [ 0.36335752  0.31615964  0.32048284]
episode: 1878000, epsilon: 0.9, loss: 0.0445548, win_rate: [ 0.36341382  0.31620039  0.32038579]
episode: 1880000, epsilon: 0.9, loss: 0.07744, win_rate: [ 0.36332597  0.31625449  0.32041954]
episode: 1882000, epsilon: 0.9, loss: 0.0353805, win_rate: [ 0.36331595  0.31614329  0.32054076]
episode: 1884000, epsilon: 0.9, loss: 0.183266, win_rate: [ 0.36329511  0.31619165  0.32051324]
episode: 1886000, epsilon: 0.9, loss: 0.123042, win_rate: [ 0.36333152  0.31619061  0.32047787]
episode: 1888000, epsilon: 0.9, loss: 0.0445418, win_rate: [ 0.36330543  0.31613557  0.32055901]
episode: 1890000, epsilon: 0.9, loss: 0.0726518, win_rate: [ 0.36326887  0.31620199  0.32052914]
episode: 1892000, epsilon: 0.9, loss: 0.374038, win_rate: [ 0.36328079  0.31615282  0.32056639]
episode: 1894000, epsilon: 0.9, loss: 0.26418, win_rate: [ 0.36330056  0.31611214  0.3205873 ]
episode: 1896000, epsilon: 0.9, loss: 0.149343, win_rate: [ 0.36332012  0.31611684  0.32056305]
episode: 1898000, epsilon: 0.9, loss: 0.220352, win_rate: [ 0.36319222  0.31626085  0.32054694]
episode: 1900000, epsilon: 0.9, loss: 0.326055, win_rate: [ 0.36312059  0.3162124   0.32066702]
episode: 1902000, epsilon: 0.9, loss: 0.213899, win_rate: [ 0.36318241  0.3162191   0.32059849]
episode: 1904000, epsilon: 0.9, loss: 0.171206, win_rate: [ 0.3631737   0.31619467  0.32063163]
episode: 1906000, epsilon: 0.9, loss: 0.0658701, win_rate: [ 0.36314448  0.31629925  0.32055627]
episode: 1908000, epsilon: 0.9, loss: 0.213199, win_rate: [ 0.36317704  0.31634896  0.32047399]
episode: 1910000, epsilon: 0.9, loss: 0.0911226, win_rate: [ 0.36311241  0.31641856  0.32046903]
episode: 1912000, epsilon: 0.9, loss: 0.295585, win_rate: [ 0.36314228  0.31634035  0.32051737]
episode: 1914000, epsilon: 0.9, loss: 0.0531822, win_rate: [ 0.36311128  0.3163462   0.32054252]
episode: 1916000, epsilon: 0.9, loss: 0.0795867, win_rate: [ 0.36309566  0.31644238  0.32046196]
episode: 1918000, epsilon: 0.9, loss: 0.071706, win_rate: [ 0.36300525  0.31654259  0.32045216]
episode: 1920000, epsilon: 0.9, loss: 0.0346108, win_rate: [ 0.36291822  0.31649018  0.3205916 ]
episode: 1922000, epsilon: 0.9, loss: 0.0286025, win_rate: [ 0.362884    0.31644324  0.32067277]
episode: 1924000, epsilon: 0.9, loss: 0.108578, win_rate: [ 0.3628698  0.3164066  0.3207236]
episode: 1926000, epsilon: 0.9, loss: 0.0697566, win_rate: [ 0.36286554  0.31634337  0.32079108]
episode: 1928000, epsilon: 0.9, loss: 0.0515931, win_rate: [ 0.36281258  0.31635876  0.32082866]
episode: 1930000, epsilon: 0.9, loss: 0.146385, win_rate: [ 0.36291778  0.31627941  0.32080281]
episode: 1932000, epsilon: 0.9, loss: 0.130552, win_rate: [ 0.36298335  0.31619117  0.32082549]
episode: 1934000, epsilon: 0.9, loss: 0.116463, win_rate: [ 0.36304588  0.31621427  0.32073985]
episode: 1936000, epsilon: 0.9, loss: 0.283969, win_rate: [ 0.3630098   0.31625149  0.3207387 ]
episode: 1938000, epsilon: 0.9, loss: 0.0865868, win_rate: [ 0.36306924  0.31617178  0.32075898]
episode: 1940000, epsilon: 0.9, loss: 0.0557935, win_rate: [ 0.3631068   0.31611414  0.32077907]
episode: 1942000, epsilon: 0.9, loss: 0.203831, win_rate: [ 0.36308744  0.31610653  0.32080603]
episode: 1944000, epsilon: 0.9, loss: 0.174596, win_rate: [ 0.36308937  0.31610368  0.32080694]
episode: 1946000, epsilon: 0.9, loss: 0.110832, win_rate: [ 0.36303292  0.3161102   0.32085688]
episode: 1948000, epsilon: 0.9, loss: 0.0461634, win_rate: [ 0.36312805  0.31606554  0.32080641]
episode: 1950000, epsilon: 0.9, loss: 0.178508, win_rate: [ 0.36310664  0.31611612  0.32077724]
episode: 1952000, epsilon: 0.9, loss: 0.0582016, win_rate: [ 0.36314299  0.31602579  0.32083122]
episode: 1954000, epsilon: 0.9, loss: 0.115294, win_rate: [ 0.36316525  0.31595233  0.32088242]
episode: 1956000, epsilon: 0.9, loss: 0.0483957, win_rate: [ 0.36312343  0.31595483  0.32092174]
episode: 1958000, epsilon: 0.9, loss: 0.0341936, win_rate: [ 0.36318419  0.31591869  0.32089712]
episode: 1960000, epsilon: 0.9, loss: 0.133591, win_rate: [ 0.36331223  0.31584219  0.32084558]
episode: 1962000, epsilon: 0.9, loss: 0.0936955, win_rate: [ 0.36331308  0.31586316  0.32082377]
episode: 1964000, epsilon: 0.9, loss: 0.112547, win_rate: [ 0.36326462  0.31587273  0.32086265]
episode: 1966000, epsilon: 0.9, loss: 0.0504683, win_rate: [ 0.36322329  0.31585768  0.32091903]
episode: 1968000, epsilon: 0.9, loss: 0.0981208, win_rate: [ 0.36313791  0.31592716  0.32093493]
episode: 1970000, epsilon: 0.9, loss: 0.0804626, win_rate: [ 0.36313288  0.31585231  0.32101481]
episode: 1972000, epsilon: 0.9, loss: 0.165911, win_rate: [ 0.3631433   0.31580453  0.32105217]
episode: 1974000, epsilon: 0.9, loss: 0.219728, win_rate: [ 0.36316897  0.31583388  0.32099715]
episode: 1976000, epsilon: 0.9, loss: 0.185247, win_rate: [ 0.3632206   0.31581279  0.32096662]
episode: 1978000, epsilon: 0.9, loss: 0.175881, win_rate: [ 0.36318053  0.31586357  0.3209559 ]
episode: 1980000, epsilon: 0.9, loss: 0.11631, win_rate: [ 0.36325546  0.31582955  0.32091499]
episode: 1982000, epsilon: 0.9, loss: 0.130356, win_rate: [ 0.36328667  0.31580013  0.3209132 ]
episode: 1984000, epsilon: 0.9, loss: 0.0408855, win_rate: [ 0.36332404  0.31574737  0.32092859]
episode: 1986000, epsilon: 0.9, loss: 0.151832, win_rate: [ 0.36329063  0.31571429  0.32099509]
episode: 1988000, epsilon: 0.9, loss: 0.0768109, win_rate: [ 0.36329152  0.315707    0.32100149]
episode: 1990000, epsilon: 0.9, loss: 0.0918728, win_rate: [ 0.36323311  0.31576117  0.32100572]
episode: 1992000, epsilon: 0.9, loss: 0.0716422, win_rate: [ 0.36327219  0.31576007  0.32096774]
episode: 1994000, epsilon: 0.9, loss: 0.0967861, win_rate: [ 0.36330254  0.31576317  0.32093429]
episode: 1996000, epsilon: 0.9, loss: 0.255496, win_rate: [ 0.36321346  0.31576207  0.32102446]
episode: 1998000, epsilon: 0.9, loss: 0.218766, win_rate: [ 0.3632334   0.31576098  0.32100562]
episode: 2000000, epsilon: 0.9, loss: 0.125588, win_rate: [ 0.36321791  0.31575783  0.32102426]
episode: 2002000, epsilon: 0.9, loss: 0.0837162, win_rate: [ 0.36323766  0.31578567  0.32097667]
episode: 2004000, epsilon: 0.9, loss: 0.0182251, win_rate: [ 0.36317911  0.31584413  0.32097676]
episode: 2006000, epsilon: 0.9, loss: 0.081153, win_rate: [ 0.36313946  0.31578743  0.32107311]
episode: 2008000, epsilon: 0.9, loss: 0.0449809, win_rate: [ 0.36309198  0.31583724  0.32107077]
episode: 2010000, epsilon: 0.9, loss: 0.0669728, win_rate: [ 0.36305505  0.3158379   0.32110705]
episode: 2012000, epsilon: 0.9, loss: 0.200649, win_rate: [ 0.36296176  0.31582035  0.32121788]
episode: 2014000, epsilon: 0.9, loss: 0.306301, win_rate: [ 0.36292565  0.3158372   0.32123715]
episode: 2016000, epsilon: 0.9, loss: 0.037215, win_rate: [ 0.36292394  0.3158238   0.32125226]
episode: 2018000, epsilon: 0.9, loss: 0.0998376, win_rate: [ 0.36288827  0.31587448  0.32123726]
episode: 2020000, epsilon: 0.9, loss: 0.118108, win_rate: [ 0.36283695  0.31588891  0.32127414]
episode: 2022000, epsilon: 0.9, loss: 0.0516394, win_rate: [ 0.3628019   0.31596867  0.32122943]
episode: 2024000, epsilon: 0.9, loss: 0.0589324, win_rate: [ 0.36277306  0.31601422  0.32121272]
episode: 2026000, epsilon: 0.9, loss: 0.440429, win_rate: [ 0.3627287   0.31605548  0.32121582]
episode: 2028000, epsilon: 0.9, loss: 0.0861787, win_rate: [ 0.36263179  0.31611013  0.32125808]
episode: 2030000, epsilon: 0.9, loss: 0.106134, win_rate: [ 0.36256686  0.31616631  0.32126684]
episode: 2032000, epsilon: 0.9, loss: 0.0953426, win_rate: [ 0.36262298  0.31612483  0.32125219]
episode: 2034000, epsilon: 0.9, loss: 0.0334055, win_rate: [ 0.36266512  0.3161011   0.32123378]
episode: 2036000, epsilon: 0.9, loss: 0.0667306, win_rate: [ 0.36266255  0.31607563  0.32126182]
episode: 2038000, epsilon: 0.9, loss: 0.126199, win_rate: [ 0.36255045  0.31613108  0.32131847]
episode: 2040000, epsilon: 0.9, loss: 0.0417682, win_rate: [ 0.36239135  0.31621099  0.32139766]
episode: 2042000, epsilon: 0.9, loss: 0.071459, win_rate: [ 0.36233836  0.31623879  0.32142285]
episode: 2044000, epsilon: 0.9, loss: 0.119106, win_rate: [ 0.36231807  0.31623979  0.32144214]
episode: 2046000, epsilon: 0.9, loss: 0.0980344, win_rate: [ 0.36227144  0.31630513  0.32142343]
episode: 2048000, epsilon: 0.9, loss: 0.105519, win_rate: [ 0.36221761  0.31633981  0.32144258]
episode: 2050000, epsilon: 0.9, loss: 0.121521, win_rate: [ 0.36220364  0.31635356  0.3214428 ]
episode: 2052000, epsilon: 0.9, loss: 0.0301801, win_rate: [ 0.36217855  0.31633726  0.32148418]
episode: 2054000, epsilon: 0.9, loss: 0.14226, win_rate: [ 0.36216856  0.31628939  0.32154205]
episode: 2056000, epsilon: 0.9, loss: 0.0683277, win_rate: [ 0.36208805  0.31627345  0.32163849]
episode: 2058000, epsilon: 0.9, loss: 0.293822, win_rate: [ 0.36211549  0.31623172  0.32165279]
episode: 2060000, epsilon: 0.9, loss: 0.0657872, win_rate: [ 0.36206897  0.31627512  0.32165591]
episode: 2062000, epsilon: 0.9, loss: 0.14516, win_rate: [ 0.36208708  0.31620246  0.32171045]
episode: 2064000, epsilon: 0.9, loss: 0.117319, win_rate: [ 0.36211605  0.31620172  0.32168223]
episode: 2066000, epsilon: 0.9, loss: 0.193634, win_rate: [ 0.36215028  0.31622834  0.32162138]
episode: 2068000, epsilon: 0.9, loss: 0.18016, win_rate: [ 0.36206251  0.31621843  0.32171906]
episode: 2070000, epsilon: 0.9, loss: 0.0634038, win_rate: [ 0.36202426  0.31621039  0.32176534]
episode: 2072000, epsilon: 0.9, loss: 0.0375553, win_rate: [ 0.36197186  0.31617175  0.3218564 ]
episode: 2074000, epsilon: 0.9, loss: 0.0981161, win_rate: [ 0.36192702  0.31617293  0.32190005]
episode: 2076000, epsilon: 0.9, loss: 0.0564145, win_rate: [ 0.36196489  0.31615798  0.32187713]
episode: 2078000, epsilon: 0.9, loss: 0.119585, win_rate: [ 0.36197751  0.31613957  0.32188292]
episode: 2080000, epsilon: 0.9, loss: 0.0647079, win_rate: [ 0.36185844  0.31620843  0.32193313]
episode: 2082000, epsilon: 0.9, loss: 0.072165, win_rate: [ 0.36176679  0.31625022  0.32198299]
episode: 2084000, epsilon: 0.9, loss: 0.0663559, win_rate: [ 0.36165107  0.31630584  0.32204309]
episode: 2086000, epsilon: 0.9, loss: 0.123386, win_rate: [ 0.36160831  0.31635404  0.32203766]
episode: 2088000, epsilon: 0.9, loss: 0.0496084, win_rate: [ 0.36163072  0.31638261  0.32198667]
episode: 2090000, epsilon: 0.9, loss: 0.093598, win_rate: [ 0.36152892  0.31641796  0.32205312]
episode: 2092000, epsilon: 0.9, loss: 0.0395417, win_rate: [ 0.36148529  0.31645133  0.32206338]
episode: 2094000, epsilon: 0.9, loss: 0.0593807, win_rate: [ 0.36141246  0.31642894  0.3221586 ]
episode: 2096000, epsilon: 0.9, loss: 0.0731772, win_rate: [ 0.3612848   0.31648625  0.32222895]
episode: 2098000, epsilon: 0.9, loss: 0.166241, win_rate: [ 0.36120627  0.31653283  0.3222609 ]
episode: 2100000, epsilon: 0.9, loss: 0.153974, win_rate: [ 0.36120556  0.31654474  0.3222497 ]
episode: 2102000, epsilon: 0.9, loss: 0.0463157, win_rate: [ 0.36118946  0.31654458  0.32226596]
episode: 2104000, epsilon: 0.9, loss: 0.0829433, win_rate: [ 0.36116152  0.31651714  0.32232134]
episode: 2106000, epsilon: 0.9, loss: 0.131953, win_rate: [ 0.36111508  0.31653238  0.32235254]
episode: 2108000, epsilon: 0.9, loss: 0.0634054, win_rate: [ 0.36105201  0.3165255   0.3224225 ]
episode: 2110000, epsilon: 0.9, loss: 0.0513609, win_rate: [ 0.36095391  0.31654398  0.32250211]
episode: 2112000, epsilon: 0.9, loss: 0.0641353, win_rate: [ 0.3609793   0.31656234  0.32245835]
episode: 2114000, epsilon: 0.9, loss: 0.0667837, win_rate: [ 0.36095925  0.31654033  0.32250042]
episode: 2116000, epsilon: 0.9, loss: 0.135232, win_rate: [ 0.36086913  0.31657363  0.32255725]
episode: 2118000, epsilon: 0.9, loss: 0.0379962, win_rate: [ 0.36076628  0.3166067   0.32262702]
episode: 2120000, epsilon: 0.9, loss: 0.0449267, win_rate: [ 0.36070729  0.31657978  0.32271293]
episode: 2122000, epsilon: 0.9, loss: 0.177008, win_rate: [ 0.36068509  0.31665233  0.32266258]
episode: 2124000, epsilon: 0.9, loss: 0.075145, win_rate: [ 0.36060861  0.31663368  0.32275771]
episode: 2126000, epsilon: 0.9, loss: 0.0583387, win_rate: [ 0.36056715  0.31669735  0.32273549]
episode: 2128000, epsilon: 0.9, loss: 0.148292, win_rate: [ 0.36048009  0.31670162  0.32281829]
episode: 2130000, epsilon: 0.9, loss: 0.310572, win_rate: [ 0.36037563  0.31673853  0.32288584]
episode: 2132000, epsilon: 0.9, loss: 0.0995076, win_rate: [ 0.3603858   0.31671008  0.32290412]
episode: 2134000, epsilon: 0.9, loss: 0.0898636, win_rate: [ 0.36039591  0.31669479  0.3229093 ]
episode: 2136000, epsilon: 0.9, loss: 0.0564059, win_rate: [ 0.36050784  0.3165858   0.32290636]
episode: 2138000, epsilon: 0.9, loss: 0.0759722, win_rate: [ 0.36042721  0.31666291  0.32290988]
episode: 2140000, epsilon: 0.9, loss: 0.27696, win_rate: [ 0.36040334  0.31669131  0.32290535]
episode: 2142000, epsilon: 0.9, loss: 0.0753155, win_rate: [ 0.36039084  0.31669069  0.32291847]
episode: 2144000, epsilon: 0.9, loss: 0.0179231, win_rate: [ 0.36033211  0.31664857  0.32301932]
episode: 2146000, epsilon: 0.9, loss: 0.0725142, win_rate: [ 0.36029763  0.31663696  0.32306541]
episode: 2148000, epsilon: 0.9, loss: 0.0200781, win_rate: [ 0.36024591  0.31666349  0.32309059]
episode: 2150000, epsilon: 0.9, loss: 0.192918, win_rate: [ 0.36023407  0.3166867   0.32307923]
episode: 2152000, epsilon: 0.9, loss: 0.0888585, win_rate: [ 0.36021756  0.31671449  0.32306795]
episode: 2154000, epsilon: 0.9, loss: 0.181318, win_rate: [ 0.36022631  0.31674367  0.32303002]
episode: 2156000, epsilon: 0.9, loss: 0.036791, win_rate: [ 0.36022717  0.31677581  0.32299702]
episode: 2158000, epsilon: 0.9, loss: 0.059742, win_rate: [ 0.36014681  0.31682024  0.32303295]
episode: 2160000, epsilon: 0.9, loss: 0.0656728, win_rate: [ 0.36008563  0.31680212  0.32311225]
episode: 2162000, epsilon: 0.9, loss: 0.0475105, win_rate: [ 0.36003259  0.31682446  0.32314295]
episode: 2164000, epsilon: 0.9, loss: 0.136057, win_rate: [ 0.36003559  0.31681263  0.32315179]
episode: 2166000, epsilon: 0.9, loss: 0.0895974, win_rate: [ 0.36001542  0.3167839   0.32320068]
episode: 2168000, epsilon: 0.9, loss: 0.147478, win_rate: [ 0.35998001  0.3167861   0.32323389]
episode: 2170000, epsilon: 0.9, loss: 0.163409, win_rate: [ 0.35994174  0.31677449  0.32328377]
episode: 2172000, epsilon: 0.9, loss: 0.0477368, win_rate: [ 0.35988232  0.3168195   0.32329818]
episode: 2174000, epsilon: 0.9, loss: 0.0357483, win_rate: [ 0.35980192  0.31683224  0.32336584]
episode: 2176000, epsilon: 0.9, loss: 0.0571089, win_rate: [ 0.35973568  0.31682819  0.32343612]
episode: 2178000, epsilon: 0.9, loss: 0.0726999, win_rate: [ 0.35967742  0.31685446  0.32346812]
episode: 2180000, epsilon: 0.9, loss: 0.0377819, win_rate: [ 0.35965575  0.31688057  0.32346369]
episode: 2182000, epsilon: 0.9, loss: 0.0220127, win_rate: [ 0.3596327   0.31684329  0.32352401]
episode: 2184000, epsilon: 0.9, loss: 0.0573208, win_rate: [ 0.35960078  0.31683626  0.32356296]
episode: 2186000, epsilon: 0.9, loss: 0.102404, win_rate: [ 0.35955559  0.31684872  0.32359569]
episode: 2188000, epsilon: 0.9, loss: 0.0493092, win_rate: [ 0.35947337  0.31692377  0.32360286]
episode: 2190000, epsilon: 0.9, loss: 0.0550277, win_rate: [ 0.35944668  0.31693887  0.32361446]
episode: 2192000, epsilon: 0.9, loss: 0.198038, win_rate: [ 0.35942904  0.31696426  0.3236067 ]
episode: 2194000, epsilon: 0.9, loss: 0.0710032, win_rate: [ 0.35933018  0.31698654  0.32368328]
episode: 2196000, epsilon: 0.9, loss: 0.178401, win_rate: [ 0.35929677  0.31701754  0.32368568]
episode: 2198000, epsilon: 0.9, loss: 0.0358399, win_rate: [ 0.35926503  0.31704248  0.32369249]
episode: 2200000, epsilon: 0.9, loss: 0.220946, win_rate: [ 0.35922468  0.31706141  0.32371391]
episode: 2202000, epsilon: 0.9, loss: 0.0476203, win_rate: [ 0.35920357  0.31704954  0.32374689]
episode: 2204000, epsilon: 0.9, loss: 0.0359026, win_rate: [ 0.35921026  0.31706105  0.32372869]
episode: 2206000, epsilon: 0.9, loss: 0.0635263, win_rate: [ 0.35922999  0.31701148  0.32375854]
episode: 2208000, epsilon: 0.9, loss: 0.10007, win_rate: [ 0.35920614  0.31701869  0.32377517]
episode: 2210000, epsilon: 0.9, loss: 0.0592156, win_rate: [ 0.35916221  0.31705908  0.32377871]
episode: 2212000, epsilon: 0.9, loss: 0.0329413, win_rate: [ 0.35910989  0.31705171  0.3238384 ]
episode: 2214000, epsilon: 0.9, loss: 0.0885201, win_rate: [ 0.35907942  0.31702283  0.32389775]
episode: 2216000, epsilon: 0.9, loss: 0.0815084, win_rate: [ 0.35905055  0.31698697  0.32396248]
episode: 2218000, epsilon: 0.9, loss: 0.0628096, win_rate: [ 0.35905326  0.31700985  0.32393688]
episode: 2220000, epsilon: 0.9, loss: 0.0940217, win_rate: [ 0.3590403   0.3170141   0.32394561]
episode: 2222000, epsilon: 0.9, loss: 0.0569249, win_rate: [ 0.35897203  0.31707511  0.32395286]
episode: 2224000, epsilon: 0.9, loss: 0.140676, win_rate: [ 0.3589537   0.31706357  0.32398273]
episode: 2226000, epsilon: 0.9, loss: 0.0302523, win_rate: [ 0.35893548  0.31704221  0.32402231]
episode: 2228000, epsilon: 0.9, loss: 0.13343, win_rate: [ 0.35892158  0.317097    0.32398142]
episode: 2230000, epsilon: 0.9, loss: 0.0500301, win_rate: [ 0.35885582  0.31713604  0.32400814]
episode: 2232000, epsilon: 0.9, loss: 0.0858408, win_rate: [ 0.35880862  0.31714266  0.32404872]
episode: 2234000, epsilon: 0.9, loss: 0.0387378, win_rate: [ 0.3587952   0.31712271  0.32408209]
episode: 2236000, epsilon: 0.9, loss: 0.165222, win_rate: [ 0.35875679  0.31709314  0.32415008]
episode: 2238000, epsilon: 0.9, loss: 0.236383, win_rate: [ 0.3586339   0.31719561  0.32417048]
episode: 2240000, epsilon: 0.9, loss: 0.0879809, win_rate: [ 0.35860584  0.31722553  0.32416863]
episode: 2242000, epsilon: 0.9, loss: 0.0664908, win_rate: [ 0.35851995  0.31728289  0.32419716]
episode: 2244000, epsilon: 0.9, loss: 0.10726, win_rate: [ 0.3584125   0.31730552  0.32428198]
episode: 2246000, epsilon: 0.9, loss: 0.285707, win_rate: [ 0.35839627  0.31733626  0.32426747]
episode: 2248000, epsilon: 0.9, loss: 0.125117, win_rate: [ 0.35831302  0.31738464  0.32430234]
episode: 2250000, epsilon: 0.9, loss: 0.089883, win_rate: [ 0.35827666  0.3173672   0.32435614]
episode: 2252000, epsilon: 0.9, loss: 0.0950126, win_rate: [ 0.35825412  0.31743021  0.32431567]
episode: 2254000, epsilon: 0.9, loss: 0.0714717, win_rate: [ 0.35820046  0.31746027  0.32433926]
episode: 2256000, epsilon: 0.9, loss: 0.105562, win_rate: [ 0.35817283  0.31750779  0.32431938]
episode: 2258000, epsilon: 0.9, loss: 0.0361784, win_rate: [ 0.35817506  0.31749021  0.32433473]
episode: 2260000, epsilon: 0.9, loss: 0.0932416, win_rate: [ 0.35812071  0.31753469  0.3243446 ]
episode: 2262000, epsilon: 0.9, loss: 0.208863, win_rate: [ 0.35811769  0.3174822   0.32440011]
episode: 2264000, epsilon: 0.9, loss: 0.0668473, win_rate: [ 0.35812274  0.31745545  0.32442181]
episode: 2266000, epsilon: 0.9, loss: 0.116842, win_rate: [ 0.35813845  0.31742082  0.32444073]
episode: 2268000, epsilon: 0.9, loss: 0.0465084, win_rate: [ 0.35816207  0.31741837  0.32441957]
episode: 2270000, epsilon: 0.9, loss: 0.0893843, win_rate: [ 0.3580965  0.3174465  0.324457 ]
episode: 2272000, epsilon: 0.9, loss: 0.0723399, win_rate: [ 0.35810288  0.31745062  0.32444651]
episode: 2274000, epsilon: 0.9, loss: 0.138351, win_rate: [ 0.35814756  0.31747322  0.32437921]
episode: 2276000, epsilon: 0.9, loss: 0.204212, win_rate: [ 0.35811948  0.3174667   0.32441382]
episode: 2278000, epsilon: 0.9, loss: 0.349262, win_rate: [ 0.3581744   0.31746284  0.32436275]
episode: 2280000, epsilon: 0.9, loss: 0.0459837, win_rate: [ 0.35818444  0.31744195  0.32437361]
episode: 2282000, epsilon: 0.9, loss: 0.095235, win_rate: [ 0.35822844  0.31739239  0.32437917]
episode: 2284000, epsilon: 0.9, loss: 0.113863, win_rate: [ 0.35824481  0.31737309  0.3243821 ]
episode: 2286000, epsilon: 0.9, loss: 0.300619, win_rate: [ 0.35829884  0.31736691  0.32433424]
episode: 2288000, epsilon: 0.9, loss: 0.118837, win_rate: [ 0.35837596  0.31738673  0.32423731]
episode: 2290000, epsilon: 0.9, loss: 0.185754, win_rate: [ 0.35843196  0.31734171  0.32422634]
episode: 2292000, epsilon: 0.9, loss: 0.369798, win_rate: [ 0.35849283  0.31734728  0.32415989]
episode: 2294000, epsilon: 0.9, loss: 0.128635, win_rate: [ 0.35856628  0.31733995  0.32409378]
episode: 2296000, epsilon: 0.9, loss: 0.0640526, win_rate: [ 0.35860851  0.31732751  0.32406399]
episode: 2298000, epsilon: 0.9, loss: 0.119931, win_rate: [ 0.35861207  0.31732795  0.32405998]
episode: 2300000, epsilon: 0.9, loss: 0.0720543, win_rate: [ 0.3585875   0.31733223  0.32408028]
episode: 2302000, epsilon: 0.9, loss: 0.0550817, win_rate: [ 0.3585758   0.31729823  0.32412597]
episode: 2304000, epsilon: 0.9, loss: 0.437503, win_rate: [ 0.35854381  0.31732418  0.32413201]
episode: 2306000, epsilon: 0.9, loss: 0.134099, win_rate: [ 0.35855892  0.31732716  0.32411392]
episode: 2308000, epsilon: 0.9, loss: 0.226469, win_rate: [ 0.35860053  0.31734531  0.32405416]
episode: 2310000, epsilon: 0.9, loss: 0.32386, win_rate: [ 0.35863562  0.31733561  0.32402878]
episode: 2312000, epsilon: 0.9, loss: 0.220806, win_rate: [ 0.35862772  0.31734609  0.32402619]
episode: 2314000, epsilon: 0.9, loss: 0.128535, win_rate: [ 0.35864624  0.31734773  0.32400603]
episode: 2316000, epsilon: 0.9, loss: 0.198497, win_rate: [ 0.35865589  0.3173882   0.32395591]
episode: 2318000, epsilon: 0.9, loss: 0.0671357, win_rate: [ 0.3586655   0.31739223  0.32394227]
episode: 2320000, epsilon: 0.9, loss: 0.267391, win_rate: [ 0.35866384  0.31742615  0.32391001]
episode: 2322000, epsilon: 0.9, loss: 0.239931, win_rate: [ 0.35872311  0.3174226   0.32385428]
episode: 2324000, epsilon: 0.9, loss: 0.219592, win_rate: [ 0.35875233  0.31741411  0.32383356]
episode: 2326000, epsilon: 0.9, loss: 0.195441, win_rate: [ 0.35881356  0.31739453  0.32379191]
episode: 2328000, epsilon: 0.9, loss: 0.174383, win_rate: [ 0.3588387   0.31738615  0.32377515]
episode: 2330000, epsilon: 0.9, loss: 0.153528, win_rate: [ 0.35883048  0.31743075  0.32373877]
episode: 2332000, epsilon: 0.9, loss: 0.0881367, win_rate: [ 0.35884195  0.31742724  0.32373081]
episode: 2334000, epsilon: 0.9, loss: 0.164874, win_rate: [ 0.35884724  0.31740537  0.3237474 ]
episode: 2336000, epsilon: 0.9, loss: 0.0420379, win_rate: [ 0.35885861  0.31738971  0.32375168]
episode: 2338000, epsilon: 0.9, loss: 0.106472, win_rate: [ 0.3588748   0.31742046  0.32370474]
episode: 2340000, epsilon: 0.9, loss: 0.149012, win_rate: [ 0.35890916  0.31739876  0.32369208]
episode: 2342000, epsilon: 0.9, loss: 0.0824185, win_rate: [ 0.35888875  0.31737596  0.32373529]
episode: 2344000, epsilon: 0.9, loss: 0.0568707, win_rate: [ 0.35890839  0.31738473  0.32370689]
episode: 2346000, epsilon: 0.9, loss: 0.0826138, win_rate: [ 0.35892792  0.31738742  0.32368466]
episode: 2348000, epsilon: 0.9, loss: 0.0532732, win_rate: [ 0.35896784  0.31739733  0.32363483]
episode: 2350000, epsilon: 0.9, loss: 0.138538, win_rate: [ 0.35901238  0.31740839  0.32357924]
episode: 2352000, epsilon: 0.9, loss: 0.0815073, win_rate: [ 0.35901954  0.31740621  0.32357425]
episode: 2354000, epsilon: 0.9, loss: 0.0779102, win_rate: [ 0.35900873  0.31740524  0.32358603]
episode: 2356000, epsilon: 0.9, loss: 0.180611, win_rate: [ 0.35900394  0.31741859  0.32357748]
episode: 2358000, epsilon: 0.9, loss: 0.150498, win_rate: [ 0.35900631  0.31744496  0.32354873]
episode: 2360000, epsilon: 0.9, loss: 0.154748, win_rate: [ 0.35896949  0.31750564  0.32352487]
episode: 2362000, epsilon: 0.9, loss: 0.121942, win_rate: [ 0.35900746  0.31749141  0.32350113]
episode: 2364000, epsilon: 0.9, loss: 0.0644087, win_rate: [ 0.35903108  0.3174808   0.32348812]
episode: 2366000, epsilon: 0.9, loss: 0.105023, win_rate: [ 0.35900978  0.31748438  0.32350584]
episode: 2368000, epsilon: 0.9, loss: 0.22181, win_rate: [ 0.35896272  0.31751499  0.32352229]
episode: 2370000, epsilon: 0.9, loss: 0.103282, win_rate: [ 0.35890766  0.31756306  0.32352927]
episode: 2372000, epsilon: 0.9, loss: 0.123044, win_rate: [ 0.35893012  0.31758633  0.32348355]
episode: 2374000, epsilon: 0.9, loss: 0.139556, win_rate: [ 0.35896298  0.31760948  0.32342754]
episode: 2376000, epsilon: 0.9, loss: 0.25153, win_rate: [ 0.35897472  0.31760457  0.32342072]
episode: 2378000, epsilon: 0.9, loss: 0.195684, win_rate: [ 0.35898059  0.31763571  0.3233837 ]
episode: 2380000, epsilon: 0.9, loss: 0.230184, win_rate: [ 0.35902354  0.31762032  0.32335614]
episode: 2382000, epsilon: 0.9, loss: 0.148513, win_rate: [ 0.35908481  0.31758764  0.32332755]
episode: 2384000, epsilon: 0.9, loss: 0.0967732, win_rate: [ 0.35913887  0.31752626  0.32333487]
episode: 2386000, epsilon: 0.9, loss: 0.494667, win_rate: [ 0.35914776  0.31751123  0.32334101]
episode: 2388000, epsilon: 0.9, loss: 0.0560135, win_rate: [ 0.35913363  0.31751925  0.32334712]
episode: 2390000, epsilon: 0.9, loss: 0.453078, win_rate: [ 0.3590932   0.31752608  0.32338072]
episode: 2392000, epsilon: 0.9, loss: 0.0994142, win_rate: [ 0.35906096  0.31755462  0.32338442]
episode: 2394000, epsilon: 0.9, loss: 0.15975, win_rate: [ 0.35906425  0.31755107  0.32338469]
episode: 2396000, epsilon: 0.9, loss: 0.0476429, win_rate: [ 0.35902653  0.3176056   0.32336787]
episode: 2398000, epsilon: 0.9, loss: 0.37116, win_rate: [ 0.35900716  0.31760309  0.32338975]
episode: 2400000, epsilon: 0.9, loss: 0.527945, win_rate: [ 0.3589482   0.31764593  0.32340587]
episode: 2402000, epsilon: 0.9, loss: 0.165785, win_rate: [ 0.35896189  0.31762411  0.323414  ]
episode: 2404000, epsilon: 0.9, loss: 0.296591, win_rate: [ 0.35896987  0.31763624  0.32339388]
episode: 2406000, epsilon: 0.9, loss: 0.168882, win_rate: [ 0.35896206  0.31763593  0.323402  ]
episode: 2408000, epsilon: 0.9, loss: 0.186222, win_rate: [ 0.35903066  0.31759182  0.32337751]
episode: 2410000, epsilon: 0.9, loss: 0.203465, win_rate: [ 0.35903844  0.31760731  0.32335425]
episode: 2412000, epsilon: 0.9, loss: 0.342121, win_rate: [ 0.35904059  0.31759141  0.323368  ]
episode: 2414000, epsilon: 0.9, loss: 0.255861, win_rate: [ 0.35906839  0.31761575  0.32331585]
episode: 2416000, epsilon: 0.9, loss: 0.0852336, win_rate: [ 0.35906156  0.31757876  0.32335968]
episode: 2418000, epsilon: 0.9, loss: 0.0675458, win_rate: [ 0.35905698  0.31760302  0.32334   ]
episode: 2420000, epsilon: 0.9, loss: 0.074182, win_rate: [ 0.35905464  0.31762053  0.32332484]
episode: 2422000, epsilon: 0.9, loss: 0.135318, win_rate: [ 0.35906005  0.31764016  0.32329979]
episode: 2424000, epsilon: 0.9, loss: 0.0548915, win_rate: [ 0.35906102  0.31764206  0.32329692]
episode: 2426000, epsilon: 0.9, loss: 0.063728, win_rate: [ 0.3590807   0.31766707  0.32325223]
episode: 2428000, epsilon: 0.9, loss: 0.170732, win_rate: [ 0.35904757  0.31771943  0.323233  ]
episode: 2430000, epsilon: 0.9, loss: 0.351069, win_rate: [ 0.35910227  0.31770361  0.32319412]
episode: 2432000, epsilon: 0.9, loss: 0.163243, win_rate: [ 0.3591622  0.3176616  0.3231762]
episode: 2434000, epsilon: 0.9, loss: 0.18284, win_rate: [ 0.35921314  0.31765688  0.32312998]
episode: 2436000, epsilon: 0.9, loss: 0.542688, win_rate: [ 0.35925188  0.31765763  0.32309049]
episode: 2438000, epsilon: 0.9, loss: 0.142264, win_rate: [ 0.35927958  0.31768988  0.32303053]
episode: 2440000, epsilon: 0.9, loss: 0.0972211, win_rate: [ 0.35936463  0.31764285  0.32299252]
episode: 2442000, epsilon: 0.9, loss: 0.0643383, win_rate: [ 0.35940171  0.31761333  0.32298496]
episode: 2444000, epsilon: 0.9, loss: 0.142358, win_rate: [ 0.35939113  0.31759905  0.32300982]
episode: 2446000, epsilon: 0.9, loss: 0.120795, win_rate: [ 0.35940752  0.31759776  0.32299472]
episode: 2448000, epsilon: 0.9, loss: 0.0905614, win_rate: [ 0.35939052  0.31759755  0.32301193]
episode: 2450000, epsilon: 0.9, loss: 0.25082, win_rate: [ 0.35942079  0.31759198  0.32298724]
episode: 2452000, epsilon: 0.9, loss: 0.587904, win_rate: [ 0.35942203  0.31760355  0.32297442]
episode: 2454000, epsilon: 0.9, loss: 0.164274, win_rate: [ 0.35945423  0.31762042  0.32292534]
episode: 2456000, epsilon: 0.9, loss: 0.411879, win_rate: [ 0.35943302  0.31766173  0.32290525]
episode: 2458000, epsilon: 0.9, loss: 0.0332759, win_rate: [ 0.35941827  0.31769967  0.32288206]
episode: 2460000, epsilon: 0.9, loss: 0.188709, win_rate: [ 0.35939404  0.3177141   0.32289186]
episode: 2462000, epsilon: 0.9, loss: 0.12944, win_rate: [ 0.35940273  0.31769353  0.32290374]
episode: 2464000, epsilon: 0.9, loss: 0.117378, win_rate: [ 0.35941562  0.31766353  0.32292085]
episode: 2466000, epsilon: 0.9, loss: 0.0685003, win_rate: [ 0.35944111  0.3176674   0.32289149]
episode: 2468000, epsilon: 0.9, loss: 0.185987, win_rate: [ 0.35946122  0.31765232  0.32288646]
episode: 2470000, epsilon: 0.9, loss: 0.0881637, win_rate: [ 0.35947706  0.3176499   0.32287304]
episode: 2472000, epsilon: 0.9, loss: 0.195112, win_rate: [ 0.35949073  0.31763387  0.32287541]
episode: 2474000, epsilon: 0.9, loss: 0.151303, win_rate: [ 0.35952734  0.31761895  0.32285371]
episode: 2476000, epsilon: 0.9, loss: 0.0899678, win_rate: [ 0.35954294  0.31760826  0.32284879]
episode: 2478000, epsilon: 0.9, loss: 0.0859148, win_rate: [ 0.359597    0.31756639  0.32283661]
episode: 2480000, epsilon: 0.9, loss: 0.0628492, win_rate: [ 0.3595916   0.31757768  0.32283072]
episode: 2482000, epsilon: 0.9, loss: 0.177395, win_rate: [ 0.35956238  0.31760552  0.32283211]
episode: 2484000, epsilon: 0.9, loss: 0.271839, win_rate: [ 0.35956328  0.3176053   0.32283142]
episode: 2486000, epsilon: 0.9, loss: 0.173625, win_rate: [ 0.35961066  0.31761541  0.32277393]
episode: 2488000, epsilon: 0.9, loss: 0.201781, win_rate: [ 0.35962692  0.31762857  0.32274451]
episode: 2490000, epsilon: 0.9, loss: 0.201062, win_rate: [ 0.3596102   0.31763447  0.32275532]
episode: 2492000, epsilon: 0.9, loss: 0.515027, win_rate: [ 0.3596459   0.31763317  0.32272093]
episode: 2494000, epsilon: 0.9, loss: 0.070646, win_rate: [ 0.35964765  0.31766773  0.32268463]
episode: 2496000, epsilon: 0.9, loss: 0.285982, win_rate: [ 0.35970357  0.31762547  0.32267096]
episode: 2498000, epsilon: 0.9, loss: 0.311496, win_rate: [ 0.35971845  0.31765582  0.32262573]
episode: 2500000, epsilon: 0.9, loss: 0.539241, win_rate: [ 0.35975059  0.31766059  0.32258882]
episode: 2502000, epsilon: 0.9, loss: 0.0951375, win_rate: [ 0.35975516  0.31769379  0.32255105]
episode: 2504000, epsilon: 0.9, loss: 0.211923, win_rate: [ 0.35979824  0.31770151  0.32250025]
episode: 2506000, epsilon: 0.9, loss: 0.1721, win_rate: [ 0.35984721  0.31768694  0.32246585]
episode: 2508000, epsilon: 0.9, loss: 0.101889, win_rate: [ 0.3598657  0.3176704  0.3224639]
episode: 2510000, epsilon: 0.9, loss: 0.11892, win_rate: [ 0.35988713  0.31766603  0.32244684]
episode: 2512000, epsilon: 0.9, loss: 0.266207, win_rate: [ 0.35990747  0.31766569  0.32242683]
episode: 2514000, epsilon: 0.9, loss: 0.104775, win_rate: [ 0.35990866  0.31769347  0.32239787]
episode: 2516000, epsilon: 0.9, loss: 0.096539, win_rate: [ 0.35991185  0.31769809  0.32239006]
episode: 2518000, epsilon: 0.9, loss: 0.135179, win_rate: [ 0.35995101  0.3176737   0.32237529]
episode: 2520000, epsilon: 0.9, loss: 0.282957, win_rate: [ 0.35997606  0.31766138  0.32236257]
episode: 2522000, epsilon: 0.9, loss: 0.203051, win_rate: [ 0.35999701  0.31766803  0.32233496]
episode: 2524000, epsilon: 0.9, loss: 0.314241, win_rate: [ 0.36000497  0.31766869  0.32232634]
episode: 2526000, epsilon: 0.9, loss: 0.0663699, win_rate: [ 0.36002182  0.31766637  0.32231181]
episode: 2528000, epsilon: 0.9, loss: 0.105806, win_rate: [ 0.3600198   0.31766604  0.32231416]
episode: 2530000, epsilon: 0.9, loss: 0.202474, win_rate: [ 0.36004149  0.31762126  0.32233725]
episode: 2532000, epsilon: 0.9, loss: 0.232697, win_rate: [ 0.36005422  0.31762693  0.32231884]
episode: 2534000, epsilon: 0.9, loss: 0.300654, win_rate: [ 0.36007478  0.31763652  0.32228869]
episode: 2536000, epsilon: 0.9, loss: 0.517826, win_rate: [ 0.36009231  0.31764215  0.32226554]
episode: 2538000, epsilon: 0.9, loss: 0.149628, win_rate: [ 0.36013819  0.31761737  0.32224444]
episode: 2540000, epsilon: 0.9, loss: 0.761278, win_rate: [ 0.36013108  0.31764551  0.32222342]
episode: 2542000, epsilon: 0.9, loss: 0.144489, win_rate: [ 0.36016109  0.31763741  0.3222015 ]
episode: 2544000, epsilon: 0.9, loss: 0.108721, win_rate: [ 0.36021046  0.31761181  0.32217773]
episode: 2546000, epsilon: 0.9, loss: 0.691901, win_rate: [ 0.36019158  0.3176291   0.32217933]
episode: 2548000, epsilon: 0.9, loss: 0.222431, win_rate: [ 0.36017471  0.31765796  0.32216733]
episode: 2550000, epsilon: 0.9, loss: 0.157987, win_rate: [ 0.36021215  0.31765669  0.32213116]
learning_rate = 0.00001
episode: 2550000, epsilon: 0.9, loss: 0, win_rate: [ 0.  1.  0.]
episode: 2552000, epsilon: 0.902296000000066, loss: 0.887622, win_rate: [ 0.36481759  0.33033483  0.30484758]
episode: 2554000, epsilon: 0.9050980000001466, loss: 0.33225, win_rate: [ 0.3644089   0.32591852  0.30967258]
episode: 2556000, epsilon: 0.907895000000227, loss: 0.251363, win_rate: [ 0.3586069   0.32577904  0.31561406]
episode: 2558000, epsilon: 0.9107010000003077, loss: 0.490338, win_rate: [ 0.36257968  0.32345957  0.31396075]
episode: 2560000, epsilon: 0.9135030000003883, loss: 0.2153, win_rate: [ 0.36546345  0.32186781  0.31266873]
episode: 2562000, epsilon: 0.9162910000004685, loss: 0.527305, win_rate: [ 0.36505291  0.32155654  0.31339055]
episode: 2564000, epsilon: 0.9190760000005486, loss: 0.380253, win_rate: [ 0.36583101  0.32033426  0.31383473]
episode: 2566000, epsilon: 0.9218800000006292, loss: 0.119594, win_rate: [ 0.36716455  0.31966752  0.31316793]
episode: 2568000, epsilon: 0.9246660000007093, loss: 0.440003, win_rate: [ 0.36809066  0.31976001  0.31214933]
episode: 2570000, epsilon: 0.9274650000007898, loss: 0.312293, win_rate: [ 0.36928154  0.31893405  0.31178441]
episode: 2572000, epsilon: 0.9302230000008691, loss: 0.235987, win_rate: [ 0.3703468   0.31766738  0.31198582]
episode: 2574000, epsilon: 0.9330130000009493, loss: 0.469618, win_rate: [ 0.37173451  0.31582017  0.31244531]
episode: 2576000, epsilon: 0.9357790000010289, loss: 0.180229, win_rate: [ 0.3737164   0.31452636  0.31175724]
episode: 2578000, epsilon: 0.9385520000011086, loss: 0.108207, win_rate: [ 0.37420092  0.31498875  0.31081033]
episode: 2580000, epsilon: 0.9413300000011885, loss: 0.0996491, win_rate: [ 0.37388754  0.31555615  0.31055631]
episode: 2582000, epsilon: 0.9441140000012685, loss: 0.10154, win_rate: [ 0.37355083  0.31583388  0.31061529]
episode: 2584000, epsilon: 0.9468800000013481, loss: 0.184527, win_rate: [ 0.3734596   0.3149025   0.31163789]
episode: 2586000, epsilon: 0.9496640000014281, loss: 0.113796, win_rate: [ 0.37371184  0.31513014  0.31115802]
episode: 2588000, epsilon: 0.95, loss: 0.243025, win_rate: [ 0.3733586   0.31509697  0.31154443]
episode: 2590000, epsilon: 0.95, loss: 0.0753302, win_rate: [ 0.37391565  0.31441714  0.31166721]
episode: 2592000, epsilon: 0.95, loss: 0.191945, win_rate: [ 0.37451489  0.31351634  0.31196876]
episode: 2594000, epsilon: 0.95, loss: 0.127587, win_rate: [ 0.37328697  0.31372014  0.31299289]
episode: 2596000, epsilon: 0.95, loss: 0.127187, win_rate: [ 0.37242669  0.31377579  0.31379753]
episode: 2598000, epsilon: 0.95, loss: 0.0640443, win_rate: [ 0.37176309  0.31428512  0.31395179]
episode: 2600000, epsilon: 0.95, loss: 0.10223, win_rate: [ 0.37143257  0.31457371  0.31399372]
episode: 2602000, epsilon: 0.95, loss: 0.302912, win_rate: [ 0.37210823  0.31435934  0.31353243]
episode: 2604000, epsilon: 0.95, loss: 0.422635, win_rate: [ 0.3726968   0.3144201   0.31288309]
episode: 2606000, epsilon: 0.95, loss: 0.213909, win_rate: [ 0.37299334  0.31431582  0.31269084]
episode: 2608000, epsilon: 0.95, loss: 0.0966619, win_rate: [ 0.37306253  0.31430493  0.31263254]
episode: 2610000, epsilon: 0.95, loss: 0.192945, win_rate: [ 0.37311045  0.31434476  0.31254479]
episode: 2612000, epsilon: 0.95, loss: 0.438287, win_rate: [ 0.37344559  0.31418848  0.31236593]
episode: 2614000, epsilon: 0.95, loss: 0.347437, win_rate: [ 0.37361916  0.31437009  0.31201075]
episode: 2616000, epsilon: 0.95, loss: 0.160285, win_rate: [ 0.37322162  0.31479826  0.31198012]
episode: 2618000, epsilon: 0.95, loss: 0.267788, win_rate: [ 0.37355333  0.31476008  0.31168659]
episode: 2620000, epsilon: 0.95, loss: 0.21512, win_rate: [ 0.37418037  0.31456693  0.3112527 ]
episode: 2622000, epsilon: 0.95, loss: 0.515683, win_rate: [ 0.37538368  0.31402342  0.31059291]
episode: 2624000, epsilon: 0.95, loss: 0.306314, win_rate: [ 0.37598141  0.31432008  0.30969852]
episode: 2626000, epsilon: 0.95, loss: 0.3078, win_rate: [ 0.37673189  0.31399587  0.30927225]
episode: 2628000, epsilon: 0.95, loss: 0.191324, win_rate: [ 0.37702081  0.31404726  0.30893194]
episode: 2630000, epsilon: 0.95, loss: 0.277246, win_rate: [ 0.37707029  0.31392108  0.30900864]
episode: 2632000, epsilon: 0.95, loss: 0.397227, win_rate: [ 0.37751979  0.31391081  0.30856941]
episode: 2634000, epsilon: 0.95, loss: 0.281449, win_rate: [ 0.3780074   0.31397245  0.30802014]
episode: 2636000, epsilon: 0.95, loss: 0.209862, win_rate: [ 0.37876304  0.31333357  0.3079034 ]
episode: 2638000, epsilon: 0.95, loss: 0.394649, win_rate: [ 0.37977978  0.3127919   0.30742832]
episode: 2640000, epsilon: 0.95, loss: 0.174992, win_rate: [ 0.38035133  0.31230764  0.30734103]
episode: 2642000, epsilon: 0.95, loss: 0.4508, win_rate: [ 0.38099586  0.31211617  0.30688797]
episode: 2644000, epsilon: 0.95, loss: 0.282224, win_rate: [ 0.38193211  0.3115286   0.30653929]
episode: 2646000, epsilon: 0.95, loss: 0.124432, win_rate: [ 0.38224602  0.31162175  0.30613223]
episode: 2648000, epsilon: 0.95, loss: 0.185114, win_rate: [ 0.38227161  0.31177233  0.30595606]
episode: 2650000, epsilon: 0.95, loss: 0.273442, win_rate: [ 0.38252617  0.31175688  0.30571694]
episode: 2652000, epsilon: 0.95, loss: 0.115005, win_rate: [ 0.38290801  0.31180086  0.30529112]
episode: 2654000, epsilon: 0.95, loss: 0.115797, win_rate: [ 0.38345785  0.31139124  0.30515091]
episode: 2656000, epsilon: 0.95, loss: 0.121879, win_rate: [ 0.38337374  0.31122348  0.30540278]
episode: 2658000, epsilon: 0.95, loss: 0.127062, win_rate: [ 0.38322793  0.31123786  0.30553421]
episode: 2660000, epsilon: 0.95, loss: 0.267072, win_rate: [ 0.38344197  0.31122444  0.30533359]
episode: 2662000, epsilon: 0.95, loss: 0.141986, win_rate: [ 0.38401443  0.31079187  0.3051937 ]
episode: 2664000, epsilon: 0.95, loss: 0.185289, win_rate: [ 0.38439136  0.31075166  0.30485697]
episode: 2666000, epsilon: 0.95, loss: 0.186698, win_rate: [ 0.38456565  0.31082491  0.30460944]
episode: 2668000, epsilon: 0.95, loss: 0.450128, win_rate: [ 0.38473403  0.3107855   0.30448047]
episode: 2670000, epsilon: 0.95, loss: 0.297058, win_rate: [ 0.38511346  0.31067241  0.30421413]
episode: 2672000, epsilon: 0.95, loss: 0.766077, win_rate: [ 0.38526733  0.31051385  0.30421882]
episode: 2674000, epsilon: 0.95, loss: 0.211789, win_rate: [ 0.38569044  0.31032008  0.30398948]
episode: 2676000, epsilon: 0.95, loss: 0.145027, win_rate: [ 0.38606043  0.31002929  0.30391029]
episode: 2678000, epsilon: 0.95, loss: 0.332892, win_rate: [ 0.38634854  0.31002102  0.30363044]
episode: 2680000, epsilon: 0.95, loss: 0.4873, win_rate: [ 0.38693549  0.30960531  0.3034592 ]
episode: 2682000, epsilon: 0.95, loss: 0.313409, win_rate: [ 0.38700464  0.30995977  0.30303558]
episode: 2684000, epsilon: 0.95, loss: 0.245766, win_rate: [ 0.38742248  0.30969172  0.3028858 ]
episode: 2686000, epsilon: 0.95, loss: 0.122358, win_rate: [ 0.3876148   0.30949772  0.30288748]
episode: 2688000, epsilon: 0.95, loss: 0.232987, win_rate: [ 0.38783777  0.30927312  0.30288911]
episode: 2690000, epsilon: 0.95, loss: 0.127682, win_rate: [ 0.38809009  0.30925493  0.30265498]
episode: 2692000, epsilon: 0.95, loss: 0.211261, win_rate: [ 0.38831417  0.30923022  0.30245562]
episode: 2694000, epsilon: 0.95, loss: 0.357715, win_rate: [ 0.38832369  0.3091923   0.30248401]
episode: 2696000, epsilon: 0.95, loss: 0.315643, win_rate: [ 0.38827816  0.30917596  0.30254587]
episode: 2698000, epsilon: 0.95, loss: 0.890412, win_rate: [ 0.38837575  0.30926818  0.30235607]
episode: 2700000, epsilon: 0.95, loss: 0.149171, win_rate: [ 0.38865074  0.30903794  0.30231132]
episode: 2702000, epsilon: 0.95, loss: 0.154567, win_rate: [ 0.38887244  0.30908349  0.30204407]
episode: 2704000, epsilon: 0.95, loss: 0.161641, win_rate: [ 0.38872475  0.30925773  0.30201752]
episode: 2706000, epsilon: 0.95, loss: 0.170755, win_rate: [ 0.38865777  0.30924802  0.30209422]
episode: 2708000, epsilon: 0.95, loss: 0.229691, win_rate: [ 0.38862412  0.30923855  0.30213733]
episode: 2710000, epsilon: 0.95, loss: 0.368382, win_rate: [ 0.38901007  0.30894182  0.30204811]
episode: 2712000, epsilon: 0.95, loss: 0.26089, win_rate: [ 0.38920747  0.30877587  0.30201665]
episode: 2714000, epsilon: 0.95, loss: 0.24603, win_rate: [ 0.38941226  0.30883958  0.30174816]
episode: 2716000, epsilon: 0.95, loss: 0.708611, win_rate: [ 0.38963018  0.30869091  0.30167891]
episode: 2718000, epsilon: 0.95, loss: 0.0997703, win_rate: [ 0.38952149  0.30873626  0.30174225]
episode: 2720000, epsilon: 0.95, loss: 1.09943, win_rate: [ 0.3897683   0.30885112  0.30138058]
episode: 2722000, epsilon: 0.95, loss: 0.402124, win_rate: [ 0.38986983  0.30875402  0.30137615]
episode: 2724000, epsilon: 0.95, loss: 0.118834, win_rate: [ 0.39016442  0.30850972  0.30132585]
episode: 2726000, epsilon: 0.95, loss: 0.261177, win_rate: [ 0.39046369  0.30845847  0.30107783]
episode: 2728000, epsilon: 0.95, loss: 0.45184, win_rate: [ 0.390745    0.30838591  0.3008691 ]
episode: 2730000, epsilon: 0.95, loss: 0.115547, win_rate: [ 0.39075894  0.30824829  0.30099277]
episode: 2732000, epsilon: 0.95, loss: 0.705942, win_rate: [ 0.39069016  0.30831699  0.30099285]
episode: 2734000, epsilon: 0.95, loss: 0.376882, win_rate: [ 0.39077505  0.30828637  0.30093858]
episode: 2736000, epsilon: 0.95, loss: 0.0835223, win_rate: [ 0.39076134  0.30827791  0.30096075]
episode: 2738000, epsilon: 0.95, loss: 0.175268, win_rate: [ 0.39106175  0.30807283  0.30086542]
episode: 2740000, epsilon: 0.95, loss: 0.131146, win_rate: [ 0.39131899  0.30799838  0.30068263]
episode: 2742000, epsilon: 0.95, loss: 0.357582, win_rate: [ 0.39155004  0.30786298  0.30058698]
episode: 2744000, epsilon: 0.95, loss: 0.691917, win_rate: [ 0.39194128  0.30760666  0.30045206]
episode: 2746000, epsilon: 0.95, loss: 0.951275, win_rate: [ 0.39203882  0.307677    0.30028418]
episode: 2748000, epsilon: 0.95, loss: 0.68881, win_rate: [ 0.39218994  0.30767016  0.3001399 ]
episode: 2750000, epsilon: 0.95, loss: 0.118453, win_rate: [ 0.39242304  0.30762846  0.2999485 ]
episode: 2752000, epsilon: 0.95, loss: 0.351353, win_rate: [ 0.39255251  0.3077163   0.29973119]
episode: 2754000, epsilon: 0.95, loss: 0.475346, win_rate: [ 0.39274317  0.30766516  0.29959167]
episode: 2756000, epsilon: 0.95, loss: 1.00133, win_rate: [ 0.39274081  0.30781889  0.29944029]
episode: 2758000, epsilon: 0.95, loss: 0.343614, win_rate: [ 0.39299811  0.30769564  0.29930625]
episode: 2760000, epsilon: 0.95, loss: 0.3343, win_rate: [ 0.39296003  0.30779377  0.29924619]
episode: 2762000, epsilon: 0.95, loss: 0.125281, win_rate: [ 0.39310192  0.30786647  0.29903161]
episode: 2764000, epsilon: 0.95, loss: 0.175751, win_rate: [ 0.39317106  0.30799856  0.29883038]
episode: 2766000, epsilon: 0.95, loss: 0.234633, win_rate: [ 0.39322966  0.30802635  0.29874399]
episode: 2768000, epsilon: 0.95, loss: 0.291845, win_rate: [ 0.3932459   0.30800776  0.29874634]
episode: 2770000, epsilon: 0.95, loss: 0.202839, win_rate: [ 0.39328003  0.30803951  0.29868046]
episode: 2772000, epsilon: 0.95, loss: 0.150783, win_rate: [ 0.39320544  0.3080887   0.29870586]
episode: 2774000, epsilon: 0.95, loss: 0.408493, win_rate: [ 0.39305182  0.3080522   0.29889599]
episode: 2776000, epsilon: 0.95, loss: 0.175026, win_rate: [ 0.39302481  0.30804288  0.29893231]
episode: 2778000, epsilon: 0.95, loss: 0.237462, win_rate: [ 0.39292372  0.30802935  0.29904693]
episode: 2780000, epsilon: 0.95, loss: 0.0495965, win_rate: [ 0.39270699  0.3082204   0.29907261]
episode: 2782000, epsilon: 0.95, loss: 0.595317, win_rate: [ 0.39269227  0.3082185   0.29908923]
episode: 2784000, epsilon: 0.95, loss: 0.410988, win_rate: [ 0.39269063  0.30825082  0.29905855]
episode: 2786000, epsilon: 0.95, loss: 0.157069, win_rate: [ 0.39265088  0.30827412  0.299075  ]
episode: 2788000, epsilon: 0.95, loss: 0.501697, win_rate: [ 0.39272524  0.30828442  0.29899034]
episode: 2790000, epsilon: 0.95, loss: 0.640196, win_rate: [ 0.39277336  0.30833622  0.29889042]
episode: 2792000, epsilon: 0.95, loss: 0.0736174, win_rate: [ 0.39282482  0.30827559  0.29889959]
episode: 2794000, epsilon: 0.95, loss: 0.368486, win_rate: [ 0.39286724  0.30818316  0.29894959]
episode: 2796000, epsilon: 0.95, loss: 0.204015, win_rate: [ 0.39297401  0.3081085   0.29891748]
episode: 2798000, epsilon: 0.95, loss: 0.153341, win_rate: [ 0.3931839   0.30792215  0.29889396]
episode: 2800000, epsilon: 0.95, loss: 0.159282, win_rate: [ 0.39332643  0.30780677  0.2988668 ]
episode: 2802000, epsilon: 0.95, loss: 0.150829, win_rate: [ 0.39345876  0.30775672  0.29878453]
episode: 2804000, epsilon: 0.95, loss: 0.188024, win_rate: [ 0.39340396  0.307798    0.29879804]
episode: 2806000, epsilon: 0.95, loss: 0.589996, win_rate: [ 0.39320159  0.30783083  0.29896758]
episode: 2808000, epsilon: 0.95, loss: 0.50735, win_rate: [ 0.39323103  0.30786315  0.29890582]
episode: 2810000, epsilon: 0.95, loss: 0.152856, win_rate: [ 0.39319464  0.30799112  0.29881424]
episode: 2812000, epsilon: 0.95, loss: 0.318983, win_rate: [ 0.39328858  0.30784615  0.29886527]
episode: 2814000, epsilon: 0.95, loss: 0.510383, win_rate: [ 0.39316139  0.30786247  0.29897614]
episode: 2816000, epsilon: 0.95, loss: 0.0910651, win_rate: [ 0.39325792  0.3078635   0.29887858]
episode: 2818000, epsilon: 0.95, loss: 0.304149, win_rate: [ 0.39324107  0.30786825  0.29889068]
episode: 2820000, epsilon: 0.95, loss: 0.583478, win_rate: [ 0.39331336  0.30788034  0.2988063 ]
episode: 2822000, epsilon: 0.95, loss: 0.163512, win_rate: [ 0.39341399  0.30784078  0.29874523]
episode: 2824000, epsilon: 0.95, loss: 0.0839384, win_rate: [ 0.39359345  0.3077288   0.29867774]
episode: 2826000, epsilon: 0.95, loss: 0.188661, win_rate: [ 0.39356741  0.30769454  0.29873805]
episode: 2828000, epsilon: 0.95, loss: 0.277757, win_rate: [ 0.393477    0.30784422  0.29867878]
episode: 2830000, epsilon: 0.95, loss: 0.590662, win_rate: [ 0.39339502  0.30785604  0.29874893]
episode: 2832000, epsilon: 0.95, loss: 0.291905, win_rate: [ 0.39342414  0.30781451  0.29876135]
episode: 2834000, epsilon: 0.95, loss: 0.381978, win_rate: [ 0.39344932  0.30774892  0.29880176]
episode: 2836000, epsilon: 0.95, loss: 0.145585, win_rate: [ 0.3934182   0.30769473  0.29888707]
episode: 2838000, epsilon: 0.95, loss: 0.221508, win_rate: [ 0.39360974  0.30761699  0.29877327]
episode: 2840000, epsilon: 0.95, loss: 0.122437, win_rate: [ 0.39352968  0.30763687  0.29883345]
episode: 2842000, epsilon: 0.95, loss: 0.199118, win_rate: [ 0.3935466  0.3076736  0.2987798]
episode: 2844000, epsilon: 0.95, loss: 0.198247, win_rate: [ 0.39361771  0.3076112   0.29877109]
episode: 2846000, epsilon: 0.95, loss: 0.381729, win_rate: [ 0.39368786  0.30754288  0.29876926]
episode: 2848000, epsilon: 0.95, loss: 0.120879, win_rate: [ 0.39360271  0.30739159  0.29900571]
episode: 2850000, epsilon: 0.95, loss: 0.131626, win_rate: [ 0.39369535  0.30725231  0.29905234]

rÖ»ÓÐ100£¬-100
episode: 2850000, epsilon: 0.9, loss: 0, win_rate: [ 1.  0.  0.]
episode: 2852000, epsilon: 0.902296000000066, loss: 0.35212, win_rate: [ 0.37731134  0.30384808  0.31884058]
episode: 2854000, epsilon: 0.9051000000001467, loss: 0.0954556, win_rate: [ 0.3824044   0.30492377  0.31267183]
episode: 2856000, epsilon: 0.9079060000002274, loss: 0.126254, win_rate: [ 0.38843526  0.29528412  0.31628062]
episode: 2858000, epsilon: 0.9107430000003089, loss: 0.422346, win_rate: [ 0.38670166  0.30521185  0.30808649]
episode: 2860000, epsilon: 0.91356000000039, loss: 0.200647, win_rate: [ 0.38946105  0.30236976  0.30816918]
episode: 2862000, epsilon: 0.916412000000472, loss: 0.132818, win_rate: [ 0.38621782  0.30422465  0.30955754]
episode: 2864000, epsilon: 0.9192160000005526, loss: 0.123212, win_rate: [ 0.38518677  0.30490679  0.30990644]
episode: 2866000, epsilon: 0.9220770000006349, loss: 0.753104, win_rate: [ 0.38622586  0.3072308   0.30654334]
episode: 2868000, epsilon: 0.9249020000007161, loss: 0.218019, win_rate: [ 0.38625632  0.30742737  0.30631632]
episode: 2870000, epsilon: 0.9276990000007965, loss: 0.6026, win_rate: [ 0.38568072  0.30833458  0.3059847 ]
episode: 2872000, epsilon: 0.9305230000008777, loss: 0.0544827, win_rate: [ 0.38725512  0.30862234  0.30412254]
episode: 2874000, epsilon: 0.9333340000009586, loss: 0.350206, win_rate: [ 0.38910879  0.30748719  0.30340402]
episode: 2876000, epsilon: 0.9361610000010399, loss: 0.159949, win_rate: [ 0.38917734  0.3085266   0.30229607]
episode: 2878000, epsilon: 0.9389890000011212, loss: 0.0692587, win_rate: [ 0.39073605  0.30806043  0.30120353]
episode: 2880000, epsilon: 0.9417880000012017, loss: 0.033563, win_rate: [ 0.39115363  0.30672311  0.30212326]
episode: 2882000, epsilon: 0.9445950000012824, loss: 0.402862, win_rate: [ 0.39067529  0.30586544  0.30345927]
episode: 2884000, epsilon: 0.9473960000013629, loss: 0.123593, win_rate: [ 0.39060616  0.30578512  0.30360872]
episode: 2886000, epsilon: 0.95, loss: 0.175677, win_rate: [ 0.39137802  0.30568595  0.30293603]
episode: 2888000, epsilon: 0.95, loss: 0.0987714, win_rate: [ 0.39180548  0.30533407  0.30286045]
episode: 2890000, epsilon: 0.95, loss: 0.103768, win_rate: [ 0.39021524  0.30584235  0.3039424 ]
episode: 2892000, epsilon: 0.95, loss: 0.116409, win_rate: [ 0.39006214  0.3059451   0.30399276]
episode: 2894000, epsilon: 0.95, loss: 0.0709723, win_rate: [ 0.39030931  0.30597032  0.30372037]
episode: 2896000, epsilon: 0.95, loss: 0.584713, win_rate: [ 0.38966544  0.30638464  0.30394991]
episode: 2898000, epsilon: 0.95, loss: 0.320018, win_rate: [ 0.39013771  0.30580613  0.30405617]
episode: 2900000, epsilon: 0.95, loss: 0.221142, win_rate: [ 0.3899922   0.30591388  0.30409392]
episode: 2902000, epsilon: 0.95, loss: 0.222141, win_rate: [ 0.39066556  0.30580181  0.30353262]
episode: 2904000, epsilon: 0.95, loss: 0.122864, win_rate: [ 0.39110387  0.30564249  0.30325364]
episode: 2906000, epsilon: 0.95, loss: 0.0750775, win_rate: [ 0.39226085  0.30556597  0.30217318]
episode: 2908000, epsilon: 0.95, loss: 0.167407, win_rate: [ 0.39275185  0.30571887  0.30152928]
episode: 2910000, epsilon: 0.95, loss: 0.249451, win_rate: [ 0.39336011  0.30557824  0.30106165]
episode: 2912000, epsilon: 0.95, loss: 0.220223, win_rate: [ 0.39315495  0.30538217  0.30146288]
episode: 2914000, epsilon: 0.95, loss: 0.0741863, win_rate: [ 0.39444696  0.30397963  0.30157341]
episode: 2916000, epsilon: 0.95, loss: 0.0441582, win_rate: [ 0.39446372  0.30370752  0.30182876]
episode: 2918000, epsilon: 0.95, loss: 0.106311, win_rate: [ 0.3944942   0.30343671  0.30206909]
episode: 2920000, epsilon: 0.95, loss: 0.0579243, win_rate: [ 0.39493722  0.30309567  0.30196711]
episode: 2922000, epsilon: 0.95, loss: 0.107152, win_rate: [ 0.39448063  0.30356523  0.30195414]
episode: 2924000, epsilon: 0.95, loss: 0.113167, win_rate: [ 0.39484602  0.3028878   0.30226619]
episode: 2926000, epsilon: 0.95, loss: 0.0243903, win_rate: [ 0.39537638  0.30260128  0.30202234]
episode: 2928000, epsilon: 0.95, loss: 0.102385, win_rate: [ 0.39563595  0.30249612  0.30186792]
episode: 2930000, epsilon: 0.95, loss: 0.111764, win_rate: [ 0.39615755  0.30189623  0.30194623]
episode: 2932000, epsilon: 0.95, loss: 0.284528, win_rate: [ 0.39600737  0.30221583  0.30177681]
episode: 2934000, epsilon: 0.95, loss: 0.0664921, win_rate: [ 0.39624528  0.30174641  0.30200831]
episode: 2936000, epsilon: 0.95, loss: 0.214716, win_rate: [ 0.39575121  0.30194998  0.30229881]
episode: 2938000, epsilon: 0.95, loss: 0.0804897, win_rate: [ 0.39560914  0.30168975  0.30270111]
episode: 2940000, epsilon: 0.95, loss: 0.14566, win_rate: [ 0.39517339  0.30176331  0.3030633 ]
episode: 2942000, epsilon: 0.95, loss: 0.0877502, win_rate: [ 0.39518049  0.3016815   0.30313801]
episode: 2944000, epsilon: 0.95, loss: 0.206966, win_rate: [ 0.39519792  0.30148615  0.30331592]
episode: 2946000, epsilon: 0.95, loss: 0.179144, win_rate: [ 0.39463131  0.30171561  0.30365309]
episode: 2948000, epsilon: 0.95, loss: 0.0266335, win_rate: [ 0.39464903  0.3014255   0.30392547]
episode: 2950000, epsilon: 0.95, loss: 0.124523, win_rate: [ 0.39534605  0.30109699  0.30355696]
episode: 2952000, epsilon: 0.95, loss: 0.176847, win_rate: [ 0.39559416  0.30088921  0.30351663]
episode: 2954000, epsilon: 0.95, loss: 0.226998, win_rate: [ 0.39583273  0.30078557  0.3033817 ]
episode: 2956000, epsilon: 0.95, loss: 0.0722165, win_rate: [ 0.3965623   0.30052547  0.30291224]
episode: 2958000, epsilon: 0.95, loss: 0.0999802, win_rate: [ 0.39672781  0.30042314  0.30284905]
episode: 2960000, epsilon: 0.95, loss: 0.0402585, win_rate: [ 0.3970873   0.30022454  0.30268816]
episode: 2962000, epsilon: 0.95, loss: 0.0841252, win_rate: [ 0.39741609  0.30022053  0.30236337]
episode: 2964000, epsilon: 0.95, loss: 0.116704, win_rate: [ 0.39754914  0.30015526  0.30229559]
episode: 2966000, epsilon: 0.95, loss: 0.145191, win_rate: [ 0.39765174  0.300075    0.30227326]
episode: 2968000, epsilon: 0.95, loss: 0.0318343, win_rate: [ 0.39786951  0.30024322  0.30188727]
episode: 2970000, epsilon: 0.95, loss: 0.060247, win_rate: [ 0.39837168  0.30010583  0.30152249]
episode: 2972000, epsilon: 0.95, loss: 0.124503, win_rate: [ 0.39872624  0.29981721  0.30145655]
episode: 2974000, epsilon: 0.95, loss: 0.36039, win_rate: [ 0.39898872  0.29962662  0.30138467]
episode: 2976000, epsilon: 0.95, loss: 0.0766514, win_rate: [ 0.39936191  0.29932302  0.30131507]
episode: 2978000, epsilon: 0.95, loss: 0.175484, win_rate: [ 0.3998875   0.29931797  0.30079453]
episode: 2980000, epsilon: 0.95, loss: 0.254144, win_rate: [ 0.39991231  0.29933616  0.30075153]
episode: 2982000, epsilon: 0.95, loss: 0.15356, win_rate: [ 0.40002727  0.29931591  0.30065681]
episode: 2984000, epsilon: 0.95, loss: 0.127987, win_rate: [ 0.40016866  0.29894553  0.30088581]
episode: 2986000, epsilon: 0.95, loss: 0.579148, win_rate: [ 0.40037206  0.29851251  0.30111543]
episode: 2988000, epsilon: 0.95, loss: 0.0744125, win_rate: [ 0.4002942   0.29828045  0.30142535]
episode: 2990000, epsilon: 0.95, loss: 0.125235, win_rate: [ 0.40010428  0.29839073  0.30150499]
episode: 2992000, epsilon: 0.95, loss: 0.197118, win_rate: [ 0.40012394  0.29832184  0.30155421]
episode: 2994000, epsilon: 0.95, loss: 0.116414, win_rate: [ 0.40033055  0.2981646   0.30150485]
episode: 2996000, epsilon: 0.95, loss: 0.2135, win_rate: [ 0.40014794  0.29817604  0.30167602]
episode: 2998000, epsilon: 0.95, loss: 0.0816322, win_rate: [ 0.40042297  0.29795069  0.30162634]
episode: 3000000, epsilon: 0.95, loss: 0.478395, win_rate: [ 0.40037733  0.29796468  0.30165799]
episode: 3002000, epsilon: 0.95, loss: 0.0943766, win_rate: [ 0.40056315  0.29791909  0.30151775]
episode: 3004000, epsilon: 0.95, loss: 0.183821, win_rate: [ 0.40074415  0.2977643   0.30149155]
episode: 3006000, epsilon: 0.95, loss: 0.239503, win_rate: [ 0.40099743  0.29772886  0.30127371]
episode: 3008000, epsilon: 0.95, loss: 0.11295, win_rate: [ 0.4010924   0.29789052  0.30101708]
episode: 3010000, epsilon: 0.95, loss: 0.361869, win_rate: [ 0.40170999  0.29774189  0.30054812]
episode: 3012000, epsilon: 0.95, loss: 0.440349, win_rate: [ 0.4021827   0.29772656  0.30009074]
episode: 3014000, epsilon: 0.95, loss: 0.140865, win_rate: [ 0.4023634   0.29765672  0.29997988]
episode: 3016000, epsilon: 0.95, loss: 0.0883075, win_rate: [ 0.40271444  0.29741387  0.29987169]
episode: 3018000, epsilon: 0.95, loss: 0.116069, win_rate: [ 0.40289641  0.29758751  0.29951607]
episode: 3020000, epsilon: 0.95, loss: 0.0569061, win_rate: [ 0.40291528  0.29751001  0.29957471]
episode: 3022000, epsilon: 0.95, loss: 0.289769, win_rate: [ 0.40329417  0.29738781  0.29931803]
episode: 3024000, epsilon: 0.95, loss: 0.627996, win_rate: [ 0.40338849  0.29752128  0.29909024]
episode: 3026000, epsilon: 0.95, loss: 0.422028, win_rate: [ 0.40353748  0.29755513  0.29890739]
episode: 3028000, epsilon: 0.95, loss: 0.297782, win_rate: [ 0.40379548  0.29739159  0.29881293]
episode: 3030000, epsilon: 0.95, loss: 0.334877, win_rate: [ 0.4042422   0.29707613  0.29868167]
episode: 3032000, epsilon: 0.95, loss: 0.13176, win_rate: [ 0.40422305  0.29702035  0.2987566 ]
episode: 3034000, epsilon: 0.95, loss: 0.0751826, win_rate: [ 0.40443258  0.29695491  0.29861251]
episode: 3036000, epsilon: 0.95, loss: 0.440611, win_rate: [ 0.40481503  0.296864    0.29832098]
episode: 3038000, epsilon: 0.95, loss: 0.136083, win_rate: [ 0.40510955  0.29687076  0.29801969]
episode: 3040000, epsilon: 0.95, loss: 0.590619, win_rate: [ 0.40528734  0.29673528  0.29797738]
episode: 3042000, epsilon: 0.95, loss: 0.338103, win_rate: [ 0.40532601  0.29662866  0.29804532]
episode: 3044000, epsilon: 0.95, loss: 0.220752, win_rate: [ 0.40536904  0.29650363  0.29812733]
episode: 3046000, epsilon: 0.95, loss: 0.209659, win_rate: [ 0.40563569  0.29633522  0.29802909]
episode: 3048000, epsilon: 0.95, loss: 0.139724, win_rate: [ 0.4060232   0.29604396  0.29793284]
episode: 3050000, epsilon: 0.95, loss: 0.398287, win_rate: [ 0.40637797  0.29586852  0.29775351]

vs cxgz self

episode: 3050000, epsilon: 0.9, loss: 0, win_rate: [ 1.  0.  0.]
episode: 3052000, epsilon: 0.9023240000000669, loss: 0.356792, win_rate: [ 0.35082459  0.33383308  0.31534233]
episode: 3054000, epsilon: 0.9051490000001481, loss: 0.158052, win_rate: [ 0.35016246  0.32916771  0.32066983]
episode: 3056000, epsilon: 0.9079550000002288, loss: 0.285282, win_rate: [ 0.35377437  0.3277787   0.31844693]
episode: 3058000, epsilon: 0.9107610000003095, loss: 0.210474, win_rate: [ 0.35595551  0.32508436  0.31896013]
episode: 3060000, epsilon: 0.9135810000003906, loss: 0.144621, win_rate: [ 0.35676432  0.32476752  0.31846815]
episode: 3062000, epsilon: 0.9164250000004723, loss: 0.212805, win_rate: [ 0.36246979  0.32088993  0.31664028]
episode: 3064000, epsilon: 0.9192460000005535, loss: 0.118537, win_rate: [ 0.36618813  0.31947718  0.31433469]
episode: 3066000, epsilon: 0.9220650000006345, loss: 0.254095, win_rate: [ 0.37035185  0.31691769  0.31273045]
episode: 3068000, epsilon: 0.9248950000007159, loss: 0.329191, win_rate: [ 0.37325704  0.313927    0.31281595]
episode: 3070000, epsilon: 0.9276990000007965, loss: 0.175907, win_rate: [ 0.37633118  0.31113444  0.31253437]
episode: 3072000, epsilon: 0.9305280000008779, loss: 0.450879, win_rate: [ 0.37730103  0.31116767  0.31153129]
episode: 3074000, epsilon: 0.9333270000009584, loss: 0.311149, win_rate: [ 0.38023416  0.31077872  0.30898713]
episode: 3076000, epsilon: 0.9361430000010393, loss: 1.28865, win_rate: [ 0.38079305  0.31098804  0.30821891]
episode: 3078000, epsilon: 0.9389310000011195, loss: 0.665161, win_rate: [ 0.38066498  0.31173887  0.30759616]
episode: 3080000, epsilon: 0.9417040000011992, loss: 1.15049, win_rate: [ 0.38142062  0.31278957  0.30578981]
episode: 3082000, epsilon: 0.94451100000128, loss: 0.262594, win_rate: [ 0.38173807  0.31330271  0.30495922]
episode: 3084000, epsilon: 0.947330000001361, loss: 0.287095, win_rate: [ 0.3826358   0.31372607  0.30363813]
episode: 3086000, epsilon: 0.95, loss: 0.784224, win_rate: [ 0.38285048  0.31482459  0.30232494]
episode: 3088000, epsilon: 0.95, loss: 1.21071, win_rate: [ 0.38422673  0.31420226  0.30157101]
episode: 3090000, epsilon: 0.95, loss: 0.317181, win_rate: [ 0.38479038  0.31444214  0.30076748]
episode: 3092000, epsilon: 0.95, loss: 0.203301, win_rate: [ 0.38527654  0.31399252  0.30073093]
episode: 3094000, epsilon: 0.95, loss: 0.0984874, win_rate: [ 0.38549124  0.31383378  0.30067498]
episode: 3096000, epsilon: 0.95, loss: 0.179117, win_rate: [ 0.38586118  0.3128845   0.30125432]
episode: 3098000, epsilon: 0.95, loss: 0.202889, win_rate: [ 0.38632528  0.3118685   0.30180621]
episode: 3100000, epsilon: 0.95, loss: 0.16447, win_rate: [ 0.38579228  0.31215376  0.30205396]
episode: 3102000, epsilon: 0.95, loss: 0.351506, win_rate: [ 0.3867041   0.31230169  0.30099421]
episode: 3104000, epsilon: 0.95, loss: 0.580899, win_rate: [ 0.3875669   0.31114239  0.30129072]
episode: 3106000, epsilon: 0.95, loss: 0.18337, win_rate: [ 0.38792164  0.31131587  0.30076249]
episode: 3108000, epsilon: 0.95, loss: 0.118663, win_rate: [ 0.38787262  0.31054637  0.30158101]
episode: 3110000, epsilon: 0.95, loss: 0.0791023, win_rate: [ 0.38782687  0.30939484  0.30277829]
episode: 3112000, epsilon: 0.95, loss: 0.218586, win_rate: [ 0.38742924  0.30920469  0.30336607]
episode: 3114000, epsilon: 0.95, loss: 0.113652, win_rate: [ 0.38685333  0.30871393  0.30443274]
episode: 3116000, epsilon: 0.95, loss: 0.059006, win_rate: [ 0.38679717  0.30785897  0.30534386]
episode: 3118000, epsilon: 0.95, loss: 0.0968436, win_rate: [ 0.38614138  0.30739254  0.30646608]
episode: 3120000, epsilon: 0.95, loss: 0.289276, win_rate: [ 0.38652305  0.30638134  0.30709561]
episode: 3122000, epsilon: 0.95, loss: 0.0909336, win_rate: [ 0.38642519  0.30592631  0.3076485 ]
episode: 3124000, epsilon: 0.95, loss: 0.105724, win_rate: [ 0.38711639  0.30614451  0.3067391 ]
episode: 3126000, epsilon: 0.95, loss: 0.464787, win_rate: [ 0.38806068  0.30608808  0.30585124]
episode: 3128000, epsilon: 0.95, loss: 0.0560692, win_rate: [ 0.38844374  0.30649607  0.30506019]
episode: 3130000, epsilon: 0.95, loss: 0.0292727, win_rate: [ 0.38798265  0.30684616  0.30517119]
episode: 3132000, epsilon: 0.95, loss: 0.298023, win_rate: [ 0.38764161  0.3071426   0.30521579]
episode: 3134000, epsilon: 0.95, loss: 0.144636, win_rate: [ 0.38715015  0.30723444  0.30561541]
episode: 3136000, epsilon: 0.95, loss: 0.35274, win_rate: [ 0.38627458  0.30746154  0.30626388]
episode: 3138000, epsilon: 0.95, loss: 0.398668, win_rate: [ 0.3862797   0.30754196  0.30617834]
episode: 3140000, epsilon: 0.95, loss: 0.283122, win_rate: [ 0.3862846   0.30786325  0.30585216]
episode: 3142000, epsilon: 0.95, loss: 0.107546, win_rate: [ 0.38612624  0.30801839  0.30585537]
episode: 3144000, epsilon: 0.95, loss: 0.0870759, win_rate: [ 0.3856342   0.30848608  0.30587972]
episode: 3146000, epsilon: 0.95, loss: 0.113818, win_rate: [ 0.3860689  0.3087572  0.3051739]
episode: 3148000, epsilon: 0.95, loss: 0.488938, win_rate: [ 0.38667973  0.30891522  0.30440506]
episode: 3150000, epsilon: 0.95, loss: 0.355036, win_rate: [ 0.38769612  0.30885691  0.30344697]
episode: 3152000, epsilon: 0.95, loss: 0.189833, win_rate: [ 0.38836874  0.30904599  0.30258527]
episode: 3154000, epsilon: 0.95, loss: 0.353438, win_rate: [ 0.3889578   0.30960279  0.30143941]
episode: 3156000, epsilon: 0.95, loss: 0.333947, win_rate: [ 0.38998689  0.30981783  0.30019528]
episode: 3158000, epsilon: 0.95, loss: 0.22085, win_rate: [ 0.39061675  0.3099138   0.29946945]
episode: 3160000, epsilon: 0.95, loss: 0.306019, win_rate: [ 0.39086917  0.31005173  0.2990791 ]
episode: 3162000, epsilon: 0.95, loss: 0.0680502, win_rate: [ 0.39130008  0.31037223  0.29832769]
episode: 3164000, epsilon: 0.95, loss: 0.310337, win_rate: [ 0.39105797  0.31100604  0.29793598]
episode: 3166000, epsilon: 0.95, loss: 1.02045, win_rate: [ 0.39147938  0.3112818   0.29723882]
episode: 3168000, epsilon: 0.95, loss: 0.162508, win_rate: [ 0.39178482  0.31154821  0.29666698]
episode: 3170000, epsilon: 0.95, loss: 0.217954, win_rate: [ 0.3925634  0.3116724  0.2957642]
episode: 3172000, epsilon: 0.95, loss: 0.635551, win_rate: [ 0.39313612  0.31180892  0.29505496]
episode: 3174000, epsilon: 0.95, loss: 0.190522, win_rate: [ 0.39355328  0.31201361  0.29443311]
episode: 3176000, epsilon: 0.95, loss: 0.161359, win_rate: [ 0.39400481  0.31229117  0.29370402]
episode: 3178000, epsilon: 0.95, loss: 0.388328, win_rate: [ 0.39415317  0.3124585   0.29338833]
episode: 3180000, epsilon: 0.95, loss: 0.221439, win_rate: [ 0.3943585   0.31258221  0.29305928]
episode: 3182000, epsilon: 0.95, loss: 0.0785915, win_rate: [ 0.39470156  0.31268703  0.29261142]
episode: 3184000, epsilon: 0.95, loss: 0.177173, win_rate: [ 0.39498213  0.31249021  0.29252767]
episode: 3186000, epsilon: 0.95, loss: 0.188309, win_rate: [ 0.39468092  0.31271829  0.29260079]
episode: 3188000, epsilon: 0.95, loss: 0.368908, win_rate: [ 0.39511308  0.31272962  0.2921573 ]
episode: 3190000, epsilon: 0.95, loss: 0.0890203, win_rate: [ 0.39499718  0.31316919  0.29183363]
episode: 3192000, epsilon: 0.95, loss: 0.345982, win_rate: [ 0.39489159  0.31302596  0.29208245]
episode: 3194000, epsilon: 0.95, loss: 1.03582, win_rate: [ 0.39509448  0.3129631   0.29194242]
episode: 3196000, epsilon: 0.95, loss: 0.830297, win_rate: [ 0.39540825  0.31312114  0.29147061]
episode: 3198000, epsilon: 0.95, loss: 0.445384, win_rate: [ 0.39601759  0.31312626  0.29085614]
episode: 3200000, epsilon: 0.95, loss: 1.33108, win_rate: [ 0.39616403  0.31342458  0.2904114 ]
episode: 3202000, epsilon: 0.95, loss: 0.625789, win_rate: [ 0.39617503  0.31376109  0.29006388]
episode: 3204000, epsilon: 0.95, loss: 0.279752, win_rate: [ 0.39646496  0.3140564   0.28947864]
episode: 3206000, epsilon: 0.95, loss: 0.677965, win_rate: [ 0.39629233  0.31456209  0.28914558]
episode: 3208000, epsilon: 0.95, loss: 0.487115, win_rate: [ 0.39645319  0.31483978  0.28870703]
episode: 3210000, epsilon: 0.95, loss: 0.501102, win_rate: [ 0.39586003  0.31564803  0.28849195]
episode: 3212000, epsilon: 0.95, loss: 0.316451, win_rate: [ 0.39567657  0.31599805  0.28832538]
episode: 3214000, epsilon: 0.95, loss: 0.105909, win_rate: [ 0.39555247  0.31621758  0.28822995]
episode: 3216000, epsilon: 0.95, loss: 0.755013, win_rate: [ 0.39590123  0.31610653  0.28799224]
episode: 3218000, epsilon: 0.95, loss: 0.0761999, win_rate: [ 0.39573574  0.31605764  0.28820662]
episode: 3220000, epsilon: 0.95, loss: 1.08494, win_rate: [ 0.39560944  0.31600991  0.28838066]
episode: 3222000, epsilon: 0.95, loss: 0.756751, win_rate: [ 0.39599188  0.31587607  0.28813205]
episode: 3224000, epsilon: 0.95, loss: 0.607813, win_rate: [ 0.39603795  0.31582577  0.28813628]
episode: 3226000, epsilon: 0.95, loss: 0.23985, win_rate: [ 0.39611139  0.31562889  0.28825973]
episode: 3228000, epsilon: 0.95, loss: 0.462458, win_rate: [ 0.39623935  0.31557126  0.28818939]
episode: 3230000, epsilon: 0.95, loss: 0.155947, win_rate: [ 0.39642558  0.31547603  0.2880984 ]
episode: 3232000, epsilon: 0.95, loss: 0.322931, win_rate: [ 0.39665167  0.31545981  0.28788853]
episode: 3234000, epsilon: 0.95, loss: 0.358559, win_rate: [ 0.39705219  0.3154222   0.28752561]
episode: 3236000, epsilon: 0.95, loss: 0.29829, win_rate: [ 0.39750862  0.31530476  0.28718663]
episode: 3238000, epsilon: 0.95, loss: 0.434371, win_rate: [ 0.39767874  0.31532811  0.28699315]
episode: 3240000, epsilon: 0.95, loss: 0.59077, win_rate: [ 0.39779791  0.31551413  0.28668796]
episode: 3242000, epsilon: 0.95, loss: 0.884634, win_rate: [ 0.39801355  0.31537336  0.28661309]
episode: 3244000, epsilon: 0.95, loss: 0.278162, win_rate: [ 0.39807011  0.31541075  0.28651914]
episode: 3246000, epsilon: 0.95, loss: 0.446636, win_rate: [ 0.39817144  0.31536574  0.28646282]
episode: 3248000, epsilon: 0.95, loss: 0.0785516, win_rate: [ 0.39813435  0.31552366  0.28634199]
episode: 3250000, epsilon: 0.95, loss: 0.133681, win_rate: [ 0.39789301  0.31571342  0.28639357]
episode: 3252000, epsilon: 0.95, loss: 0.10896, win_rate: [ 0.39770595  0.31588457  0.28640947]
episode: 3254000, epsilon: 0.95, loss: 0.0833365, win_rate: [ 0.39729707  0.31570924  0.28699369]
episode: 3256000, epsilon: 0.95, loss: 0.162044, win_rate: [ 0.39715341  0.31568293  0.28716365]
episode: 3258000, epsilon: 0.95, loss: 0.256189, win_rate: [ 0.3970029   0.31563791  0.2873592 ]
episode: 3260000, epsilon: 0.95, loss: 0.111084, win_rate: [ 0.39709811  0.31545564  0.28744625]
episode: 3262000, epsilon: 0.95, loss: 0.156901, win_rate: [ 0.3974368  0.3152721  0.2872911]
episode: 3264000, epsilon: 0.95, loss: 0.166274, win_rate: [ 0.39754954  0.31524152  0.28720894]
episode: 3266000, epsilon: 0.95, loss: 0.0916997, win_rate: [ 0.39768797  0.31531798  0.28699404]
episode: 3268000, epsilon: 0.95, loss: 0.265295, win_rate: [ 0.39792937  0.31524167  0.28682896]
episode: 3270000, epsilon: 0.95, loss: 0.574872, win_rate: [ 0.39802092  0.31505766  0.28692142]
episode: 3272000, epsilon: 0.95, loss: 0.236619, win_rate: [ 0.39823244  0.314922    0.28684555]
episode: 3274000, epsilon: 0.95, loss: 0.297531, win_rate: [ 0.39856518  0.31478431  0.28665051]
episode: 3276000, epsilon: 0.95, loss: 0.259142, win_rate: [ 0.39866195  0.3146402   0.28669785]
episode: 3278000, epsilon: 0.95, loss: 0.442417, win_rate: [ 0.39892369  0.31445476  0.28662155]
episode: 3280000, epsilon: 0.95, loss: 0.115972, win_rate: [ 0.39933305  0.31424211  0.28642484]
episode: 3282000, epsilon: 0.95, loss: 0.106646, win_rate: [ 0.39977414  0.31421416  0.2860117 ]
episode: 3284000, epsilon: 0.95, loss: 0.183198, win_rate: [ 0.40002393  0.3142337   0.28574237]
episode: 3286000, epsilon: 0.95, loss: 0.710396, win_rate: [ 0.40029915  0.31433341  0.28536743]
episode: 3288000, epsilon: 0.95, loss: 0.454794, win_rate: [ 0.40010336  0.3145617   0.28533494]
episode: 3290000, epsilon: 0.95, loss: 0.490092, win_rate: [ 0.39906917  0.31536535  0.28556548]
episode: 3292000, epsilon: 0.95, loss: 0.495297, win_rate: [ 0.39806034  0.31633754  0.28560213]
episode: 3294000, epsilon: 0.95, loss: 0.82483, win_rate: [ 0.39722378  0.31713804  0.28563817]
episode: 3296000, epsilon: 0.95, loss: 1.02413, win_rate: [ 0.39625855  0.31821415  0.28552729]
episode: 3298000, epsilon: 0.95, loss: 0.222226, win_rate: [ 0.39534921  0.31916404  0.28548675]
episode: 3300000, epsilon: 0.95, loss: 0.121787, win_rate: [ 0.39500242  0.31939472  0.28560286]
episode: 3302000, epsilon: 0.95, loss: 0.108638, win_rate: [ 0.39501431  0.31940746  0.28557823]
episode: 3304000, epsilon: 0.95, loss: 0.197468, win_rate: [ 0.39509293  0.31930977  0.2855973 ]
episode: 3306000, epsilon: 0.95, loss: 0.371732, win_rate: [ 0.39498283  0.31929172  0.28572545]
episode: 3308000, epsilon: 0.95, loss: 0.106133, win_rate: [ 0.39490157  0.3192507   0.28584773]
episode: 3310000, epsilon: 0.95, loss: 0.0169511, win_rate: [ 0.39494848  0.3191757   0.28587582]
episode: 3312000, epsilon: 0.95, loss: 0.207717, win_rate: [ 0.39480002  0.31900642  0.28619356]
episode: 3314000, epsilon: 0.95, loss: 0.03927, win_rate: [ 0.39485078  0.31877152  0.2863777 ]
episode: 3316000, epsilon: 0.95, loss: 0.0341729, win_rate: [ 0.39481055  0.31858527  0.28660419]
episode: 3318000, epsilon: 0.95, loss: 0.0267265, win_rate: [ 0.39466644  0.31847642  0.28685714]
episode: 3320000, epsilon: 0.95, loss: 0.0481421, win_rate: [ 0.39449484  0.31846549  0.28703968]
episode: 3322000, epsilon: 0.95, loss: 0.0161505, win_rate: [ 0.39441767  0.31831133  0.287271  ]
episode: 3324000, epsilon: 0.95, loss: 0.029523, win_rate: [ 0.3941336   0.31830906  0.28755734]
episode: 3326000, epsilon: 0.95, loss: 0.105369, win_rate: [ 0.39379205  0.31827059  0.28793736]
episode: 3328000, epsilon: 0.95, loss: 0.309872, win_rate: [ 0.39345902  0.31822188  0.28831911]
episode: 3330000, epsilon: 0.95, loss: 0.094965, win_rate: [ 0.39345217  0.31817029  0.28837754]
episode: 3332000, epsilon: 0.95, loss: 0.267482, win_rate: [ 0.3933213   0.31819745  0.28848125]
episode: 3334000, epsilon: 0.95, loss: 0.0510928, win_rate: [ 0.39317467  0.31817141  0.28865391]
episode: 3336000, epsilon: 0.95, loss: 0.0573661, win_rate: [ 0.39316296  0.31821567  0.28862137]
episode: 3338000, epsilon: 0.95, loss: 0.0497296, win_rate: [ 0.39313058  0.31839126  0.28847817]
episode: 3340000, epsilon: 0.95, loss: 0.335521, win_rate: [ 0.39297796  0.31853338  0.28848866]
episode: 3342000, epsilon: 0.95, loss: 0.0941293, win_rate: [ 0.39292674  0.3186222   0.28845107]
episode: 3344000, epsilon: 0.95, loss: 0.316371, win_rate: [ 0.39303608  0.31850912  0.2884548 ]
episode: 3346000, epsilon: 0.95, loss: 0.0430629, win_rate: [ 0.39306624  0.31855298  0.28838078]
episode: 3348000, epsilon: 0.95, loss: 0.074684, win_rate: [ 0.39299533  0.31853249  0.28847219]
episode: 3350000, epsilon: 0.95, loss: 0.0763739, win_rate: [ 0.39293536  0.31858227  0.28848237]

vs self cxgz
episode: 3350000, epsilon: 0.9, loss: 0, win_rate: [ 1.  0.  0.]
episode: 3352000, epsilon: 0.9024050000000692, loss: 0.0810784, win_rate: [ 0.3808096   0.31284358  0.30634683]
episode: 3354000, epsilon: 0.9052770000001518, loss: 0.065006, win_rate: [ 0.38690327  0.29717571  0.31592102]
episode: 3356000, epsilon: 0.908172000000235, loss: 0.0546308, win_rate: [ 0.38410265  0.29511748  0.32077987]
episode: 3358000, epsilon: 0.9110550000003179, loss: 0.0565547, win_rate: [ 0.38582677  0.2968379   0.31733533]
episode: 3360000, epsilon: 0.9139610000004015, loss: 0.0520461, win_rate: [ 0.38616138  0.30126987  0.31256874]
episode: 3362000, epsilon: 0.9168520000004846, loss: 0.066466, win_rate: [ 0.38730106  0.29947504  0.3132239 ]
episode: 3364000, epsilon: 0.9197430000005677, loss: 0.0939109, win_rate: [ 0.38918649  0.2976216   0.31319191]
episode: 3366000, epsilon: 0.9226250000006506, loss: 0.114554, win_rate: [ 0.3919755   0.29498156  0.31304293]
episode: 3368000, epsilon: 0.9254740000007325, loss: 0.030207, win_rate: [ 0.39220043  0.29292817  0.3148714 ]
episode: 3370000, epsilon: 0.9283320000008147, loss: 0.17055, win_rate: [ 0.39623019  0.29088546  0.31288436]
episode: 3372000, epsilon: 0.9312110000008975, loss: 0.0568116, win_rate: [ 0.39648198  0.29148675  0.31203127]
episode: 3374000, epsilon: 0.9340860000009802, loss: 0.0970796, win_rate: [ 0.39865006  0.28961293  0.31173701]
episode: 3376000, epsilon: 0.9369470000010625, loss: 0.0946604, win_rate: [ 0.3995231   0.28864274  0.31183416]
episode: 3378000, epsilon: 0.9398270000011453, loss: 0.0449225, win_rate: [ 0.40127138  0.28738259  0.31134602]
episode: 3380000, epsilon: 0.9427000000012279, loss: 0.105175, win_rate: [ 0.40241992  0.28795707  0.30962301]
episode: 3382000, epsilon: 0.945592000001311, loss: 0.103425, win_rate: [ 0.40308115  0.28736602  0.30955283]
episode: 3384000, epsilon: 0.9484740000013939, loss: 0.0780947, win_rate: [ 0.40487044  0.28593277  0.30919679]
episode: 3386000, epsilon: 0.95, loss: 0.132331, win_rate: [ 0.40532207  0.28474209  0.30993584]
episode: 3388000, epsilon: 0.95, loss: 0.216668, win_rate: [ 0.40548933  0.28536091  0.30914976]
episode: 3390000, epsilon: 0.95, loss: 0.222027, win_rate: [ 0.40616485  0.28461788  0.30921727]
episode: 3392000, epsilon: 0.95, loss: 0.110283, win_rate: [ 0.4072284   0.28392181  0.30884979]
episode: 3394000, epsilon: 0.95, loss: 0.170349, win_rate: [ 0.407968    0.28231177  0.30972023]
episode: 3396000, epsilon: 0.95, loss: 0.2929, win_rate: [ 0.40916502  0.28099389  0.30984109]
episode: 3398000, epsilon: 0.95, loss: 0.0885429, win_rate: [ 0.40915814  0.28132747  0.30951439]
episode: 3400000, epsilon: 0.95, loss: 0.161272, win_rate: [ 0.4098318   0.28115438  0.30901382]
episode: 3402000, epsilon: 0.95, loss: 0.124588, win_rate: [ 0.41001135  0.28130228  0.30868637]
episode: 3404000, epsilon: 0.95, loss: 0.064627, win_rate: [ 0.41012203  0.28053184  0.30934612]
episode: 3406000, epsilon: 0.95, loss: 0.180042, win_rate: [ 0.41092123  0.27994143  0.30913734]
episode: 3408000, epsilon: 0.95, loss: 0.186747, win_rate: [ 0.41164808  0.27940898  0.30894295]
episode: 3410000, epsilon: 0.95, loss: 0.190985, win_rate: [ 0.41252646  0.27906202  0.30841153]
episode: 3412000, epsilon: 0.95, loss: 0.248493, win_rate: [ 0.41328366  0.27893098  0.30778536]
episode: 3414000, epsilon: 0.95, loss: 0.183953, win_rate: [ 0.41310292  0.27832378  0.3085733 ]
episode: 3416000, epsilon: 0.95, loss: 0.0522808, win_rate: [ 0.41350889  0.27795033  0.30854078]
episode: 3418000, epsilon: 0.95, loss: 0.129773, win_rate: [ 0.41362627  0.27808415  0.30828958]
episode: 3420000, epsilon: 0.95, loss: 0.234202, win_rate: [ 0.41426551  0.2781246   0.30760989]
episode: 3422000, epsilon: 0.95, loss: 0.208663, win_rate: [ 0.41439702  0.27785725  0.30774573]
episode: 3424000, epsilon: 0.95, loss: 0.174379, win_rate: [ 0.41499439  0.2772395   0.30776611]
episode: 3426000, epsilon: 0.95, loss: 0.343338, win_rate: [ 0.41473139  0.27739109  0.30787753]
episode: 3428000, epsilon: 0.95, loss: 0.138028, win_rate: [ 0.41408443  0.27794515  0.30797041]
episode: 3430000, epsilon: 0.95, loss: 0.129232, win_rate: [ 0.41480731  0.27728403  0.30790865]
episode: 3432000, epsilon: 0.95, loss: 0.136292, win_rate: [ 0.41478762  0.27750881  0.30770356]
episode: 3434000, epsilon: 0.95, loss: 0.740916, win_rate: [ 0.41522125  0.27709194  0.30768681]
episode: 3436000, epsilon: 0.95, loss: 0.116247, win_rate: [ 0.41556494  0.27695027  0.3074848 ]
episode: 3438000, epsilon: 0.95, loss: 0.210872, win_rate: [ 0.41593846  0.27665595  0.3074056 ]
episode: 3440000, epsilon: 0.95, loss: 0.0430665, win_rate: [ 0.4162176   0.27623026  0.30755214]
episode: 3442000, epsilon: 0.95, loss: 0.120625, win_rate: [ 0.41694112  0.27584483  0.30721405]
episode: 3444000, epsilon: 0.95, loss: 0.147355, win_rate: [ 0.41709131  0.27557154  0.30733716]
episode: 3446000, epsilon: 0.95, loss: 0.0924376, win_rate: [ 0.41738107  0.27539296  0.30722597]
episode: 3448000, epsilon: 0.95, loss: 0.168193, win_rate: [ 0.41784268  0.27557882  0.3065785 ]
episode: 3450000, epsilon: 0.95, loss: 0.447283, win_rate: [ 0.41758582  0.27574724  0.30666693]
episode: 3452000, epsilon: 0.95, loss: 0.158062, win_rate: [ 0.41781943  0.27546789  0.30671268]
episode: 3454000, epsilon: 0.95, loss: 0.211727, win_rate: [ 0.41779406  0.27562235  0.30658359]
episode: 3456000, epsilon: 0.95, loss: 0.36154, win_rate: [ 0.41793002  0.2754125   0.30665748]
episode: 3458000, epsilon: 0.95, loss: 0.0997839, win_rate: [ 0.41842205  0.27526597  0.30631198]
episode: 3460000, epsilon: 0.95, loss: 0.261303, win_rate: [ 0.41855074  0.27512477  0.30632449]
episode: 3462000, epsilon: 0.95, loss: 0.236665, win_rate: [ 0.41776413  0.27505112  0.30718476]
episode: 3464000, epsilon: 0.95, loss: 0.216566, win_rate: [ 0.4171016   0.27508531  0.30781309]
episode: 3466000, epsilon: 0.95, loss: 1.34569, win_rate: [ 0.41679813  0.2754028   0.30779907]
episode: 3468000, epsilon: 0.95, loss: 1.13172, win_rate: [ 0.41673376  0.27560783  0.30765841]
episode: 3470000, epsilon: 0.95, loss: 0.21503, win_rate: [ 0.41661319  0.27555604  0.30783077]
episode: 3472000, epsilon: 0.95, loss: 0.570864, win_rate: [ 0.41690642  0.27533381  0.30775977]
episode: 3474000, epsilon: 0.95, loss: 0.169456, win_rate: [ 0.41697244  0.27523165  0.3077959 ]
episode: 3476000, epsilon: 0.95, loss: 0.327956, win_rate: [ 0.41677447  0.27510099  0.30812454]
episode: 3478000, epsilon: 0.95, loss: 0.271367, win_rate: [ 0.41713737  0.27500566  0.30785697]
episode: 3480000, epsilon: 0.95, loss: 1.80193, win_rate: [ 0.41742756  0.27495173  0.30762071]
episode: 3482000, epsilon: 0.95, loss: 1.47183, win_rate: [ 0.41745138  0.27527064  0.30727798]
episode: 3484000, epsilon: 0.95, loss: 0.233639, win_rate: [ 0.41738494  0.27529645  0.3073186 ]
episode: 3486000, epsilon: 0.95, loss: 0.86595, win_rate: [ 0.41733517  0.27540974  0.30725509]
episode: 3488000, epsilon: 0.95, loss: 1.28911, win_rate: [ 0.4170477   0.27533858  0.30761371]
episode: 3490000, epsilon: 0.95, loss: 0.388391, win_rate: [ 0.4166756   0.27521946  0.30810494]
episode: 3492000, epsilon: 0.95, loss: 0.181226, win_rate: [ 0.41675059  0.27503327  0.30821614]
episode: 3494000, epsilon: 0.95, loss: 0.386555, win_rate: [ 0.41678183  0.27520642  0.30801175]
episode: 3496000, epsilon: 0.95, loss: 0.318833, win_rate: [ 0.4169081   0.27529948  0.30779241]
episode: 3498000, epsilon: 0.95, loss: 1.01888, win_rate: [ 0.41703097  0.27531571  0.30765333]
episode: 3500000, epsilon: 0.95, loss: 0.205804, win_rate: [ 0.41746388  0.27486483  0.30767128]


vs cxgz cxgz Ëæ»úÑ¡Ôñplayer¿ª¾Ö
episode: 3500000, epsilon: 0.9, loss: 0, win_rate: [ 0.  1.  0.]
episode: 3502000, epsilon: 0.9023950000000689, loss: 0.806798, win_rate: [ 0.33183408  0.33583208  0.33233383]
episode: 3504000, epsilon: 0.9052950000001523, loss: 1.07571, win_rate: [ 0.32916771  0.33991502  0.33091727]
episode: 3506000, epsilon: 0.9081620000002347, loss: 0.12572, win_rate: [ 0.32911181  0.33644393  0.33444426]
episode: 3508000, epsilon: 0.9110880000003189, loss: 0.115997, win_rate: [ 0.3312086   0.33308336  0.33570804]
episode: 3510000, epsilon: 0.9140070000004028, loss: 0.424768, win_rate: [ 0.33136686  0.33166683  0.3369663 ]
episode: 3512000, epsilon: 0.9169090000004863, loss: 0.106613, win_rate: [ 0.32688943  0.33230564  0.34080493]
episode: 3514000, epsilon: 0.9198180000005699, loss: 0.161545, win_rate: [ 0.32590529  0.33333333  0.34076137]
episode: 3516000, epsilon: 0.9227160000006532, loss: 0.17353, win_rate: [ 0.32547966  0.33304168  0.34147866]
episode: 3518000, epsilon: 0.9255730000007354, loss: 0.178525, win_rate: [ 0.32359313  0.33292595  0.34348092]
episode: 3520000, epsilon: 0.9284840000008191, loss: 0.6951, win_rate: [ 0.32473376  0.33213339  0.34313284]
episode: 3522000, epsilon: 0.9314100000009032, loss: 0.321101, win_rate: [ 0.32562156  0.33125767  0.34312077]
episode: 3524000, epsilon: 0.9343010000009864, loss: 0.648478, win_rate: [ 0.32627807  0.33094454  0.34277738]
episode: 3526000, epsilon: 0.9371940000010696, loss: 0.508862, win_rate: [ 0.32679512  0.33179493  0.34140995]
episode: 3528000, epsilon: 0.9400890000011528, loss: 0.557298, win_rate: [ 0.32777401  0.33202386  0.34020214]
episode: 3530000, epsilon: 0.9430070000012367, loss: 0.196167, win_rate: [ 0.32865571  0.33158895  0.33975534]
episode: 3532000, epsilon: 0.9459170000013204, loss: 0.363737, win_rate: [ 0.3296147   0.33123965  0.33914565]
episode: 3534000, epsilon: 0.9488410000014045, loss: 0.105782, win_rate: [ 0.33057851  0.33069616  0.33872533]
episode: 3536000, epsilon: 0.95, loss: 0.102661, win_rate: [ 0.33174078  0.33096303  0.33729619]
episode: 3538000, epsilon: 0.95, loss: 0.153373, win_rate: [ 0.33175443  0.33067551  0.33757006]
episode: 3540000, epsilon: 0.95, loss: 0.213704, win_rate: [ 0.33321667  0.33021674  0.33656659]
episode: 3542000, epsilon: 0.95, loss: 0.925329, win_rate: [ 0.33351587  0.330135    0.33634913]
episode: 3544000, epsilon: 0.95, loss: 0.259469, win_rate: [ 0.33403786  0.32994705  0.33601509]
episode: 3546000, epsilon: 0.95, loss: 0.0748213, win_rate: [ 0.33507967  0.32903633  0.335884  ]
episode: 3548000, epsilon: 0.95, loss: 0.488193, win_rate: [ 0.33584717  0.32903481  0.33511802]
episode: 3550000, epsilon: 0.95, loss: 0.335258, win_rate: [ 0.33647327  0.32949341  0.33403332]
episode: 3552000, epsilon: 0.95, loss: 0.264187, win_rate: [ 0.3371089   0.32924367  0.33364743]
episode: 3554000, epsilon: 0.95, loss: 0.776677, win_rate: [ 0.33814189  0.32954945  0.33230866]
episode: 3556000, epsilon: 0.95, loss: 0.137488, win_rate: [ 0.33870824  0.33027982  0.33101195]
episode: 3558000, epsilon: 0.95, loss: 0.291691, win_rate: [ 0.33906312  0.33058051  0.33035637]
episode: 3560000, epsilon: 0.95, loss: 0.34657, win_rate: [ 0.33891102  0.33076115  0.33032783]
episode: 3562000, epsilon: 0.95, loss: 0.813504, win_rate: [ 0.33844615  0.33113982  0.33041403]
episode: 3564000, epsilon: 0.95, loss: 0.328314, win_rate: [ 0.33871346  0.33113545  0.33015109]
episode: 3566000, epsilon: 0.95, loss: 0.394356, win_rate: [ 0.33929789  0.33091923  0.32978288]
episode: 3568000, epsilon: 0.95, loss: 0.508164, win_rate: [ 0.33970089  0.3309216   0.32937751]
episode: 3570000, epsilon: 0.95, loss: 0.240494, win_rate: [ 0.33920944  0.33149526  0.3292953 ]
episode: 3572000, epsilon: 0.95, loss: 0.421952, win_rate: [ 0.33871752  0.33181484  0.32946765]
episode: 3574000, epsilon: 0.95, loss: 0.369345, win_rate: [ 0.33880623  0.33176579  0.32942798]
episode: 3576000, epsilon: 0.95, loss: 0.117256, win_rate: [ 0.33840344  0.33189037  0.32970619]
episode: 3578000, epsilon: 0.95, loss: 0.373269, win_rate: [ 0.33838028  0.33162395  0.32999577]
episode: 3580000, epsilon: 0.95, loss: 0.366827, win_rate: [ 0.33843327  0.33110836  0.33045837]
episode: 3582000, epsilon: 0.95, loss: 0.668521, win_rate: [ 0.33873977  0.33104474  0.33021549]
episode: 3584000, epsilon: 0.95, loss: 0.360169, win_rate: [ 0.33879359  0.33101987  0.33018655]
episode: 3586000, epsilon: 0.95, loss: 0.216601, win_rate: [ 0.33920536  0.33061243  0.33018221]
episode: 3588000, epsilon: 0.95, loss: 0.0305391, win_rate: [ 0.33930296  0.3304167   0.33028034]
episode: 3590000, epsilon: 0.95, loss: 0.137689, win_rate: [ 0.33928512  0.32984078  0.3308741 ]
episode: 3592000, epsilon: 0.95, loss: 0.100704, win_rate: [ 0.33933327  0.33003989  0.33062684]
episode: 3594000, epsilon: 0.95, loss: 0.156848, win_rate: [ 0.3396666   0.33008159  0.33025181]
episode: 3596000, epsilon: 0.95, loss: 0.336359, win_rate: [ 0.34015271  0.32971532  0.33013198]
episode: 3598000, epsilon: 0.95, loss: 0.14477, win_rate: [ 0.34062918  0.32915991  0.33021092]
episode: 3600000, epsilon: 0.95, loss: 0.355051, win_rate: [ 0.34089659  0.32909671  0.3300067 ]

vs combne combine


episode: 3600000, epsilon: 0.8, loss: 0, win_rate: [ 1.  0.  0.]
episode: 3602000, epsilon: 0.80243200000007, loss: 0.191315, win_rate: [ 0.27136432  0.35482259  0.37381309]
episode: 3604000, epsilon: 0.805355000000154, loss: 0.238859, win_rate: [ 0.27168208  0.35616096  0.37215696]
episode: 3606000, epsilon: 0.808310000000239, loss: 0.220396, win_rate: [ 0.26678887  0.3577737   0.37543743]
episode: 3608000, epsilon: 0.8113140000003254, loss: 0.454413, win_rate: [ 0.26671666  0.35895513  0.37432821]
episode: 3610000, epsilon: 0.8142840000004108, loss: 0.151817, win_rate: [ 0.26877312  0.35886411  0.37236276]
episode: 3612000, epsilon: 0.8172290000004955, loss: 0.263051, win_rate: [ 0.2728106   0.35438713  0.37280227]
episode: 3614000, epsilon: 0.8201860000005805, loss: 0.168866, win_rate: [ 0.27169488  0.35654596  0.37175916]
episode: 3616000, epsilon: 0.8231480000006657, loss: 1.13757, win_rate: [ 0.27235798  0.35585276  0.37178926]
episode: 3618000, epsilon: 0.8261120000007509, loss: 0.223401, win_rate: [ 0.27376257  0.35514694  0.37109049]
episode: 3620000, epsilon: 0.8290850000008364, loss: 0.531205, win_rate: [ 0.27533623  0.35448228  0.37018149]
episode: 3622000, epsilon: 0.8320540000009218, loss: 0.164576, win_rate: [ 0.27580565  0.35439298  0.36980137]
episode: 3624000, epsilon: 0.8350390000010076, loss: 0.185107, win_rate: [ 0.27623849  0.35419358  0.36956793]
episode: 3626000, epsilon: 0.8380000000010928, loss: 0.305293, win_rate: [ 0.27629707  0.35425561  0.36944733]
episode: 3628000, epsilon: 0.8409340000011771, loss: 1.30008, win_rate: [ 0.27695439  0.35298739  0.37005821]
episode: 3630000, epsilon: 0.8438920000012622, loss: 0.318573, win_rate: [ 0.27815739  0.35265491  0.36918769]
episode: 3632000, epsilon: 0.8468200000013464, loss: 0.209411, win_rate: [ 0.27911628  0.351864    0.36901972]
episode: 3634000, epsilon: 0.8497740000014313, loss: 0.109764, win_rate: [ 0.28066822  0.35040146  0.36893033]
episode: 3636000, epsilon: 0.8527450000015168, loss: 0.41571, win_rate: [ 0.28199217  0.3488792   0.36912864]
episode: 3638000, epsilon: 0.8557160000016022, loss: 0.265145, win_rate: [ 0.28251888  0.34841188  0.36906924]
episode: 3640000, epsilon: 0.8586600000016869, loss: 0.942318, win_rate: [ 0.28344291  0.3481663   0.36839079]
episode: 3642000, epsilon: 0.8615740000017706, loss: 0.220951, win_rate: [ 0.28463608  0.34739649  0.36796743]
episode: 3644000, epsilon: 0.8645290000018556, loss: 0.423382, win_rate: [ 0.28485716  0.34696939  0.36817345]
episode: 3646000, epsilon: 0.8674610000019399, loss: 0.667183, win_rate: [ 0.28568944  0.34660116  0.3677094 ]
episode: 3648000, epsilon: 0.8704220000020251, loss: 0.942241, win_rate: [ 0.28645237  0.34622195  0.36732568]
episode: 3650000, epsilon: 0.8733660000021097, loss: 1.15025, win_rate: [ 0.28703426  0.34689306  0.36607268]
episode: 3652000, epsilon: 0.8763200000021947, loss: 0.493065, win_rate: [ 0.28764831  0.34720486  0.36514682]
episode: 3654000, epsilon: 0.8792710000022795, loss: 1.58658, win_rate: [ 0.28888354  0.34645655  0.36465991]
episode: 3656000, epsilon: 0.8822170000023642, loss: 0.259723, win_rate: [ 0.28969126  0.34602954  0.36427921]
episode: 3658000, epsilon: 0.8851600000024489, loss: 0.284277, win_rate: [ 0.29102947  0.34594231  0.36302822]
episode: 3660000, epsilon: 0.8880910000025332, loss: 1.92923, win_rate: [ 0.29131181  0.34607757  0.36261062]
episode: 3662000, epsilon: 0.8910010000026168, loss: 1.10307, win_rate: [ 0.29178562  0.34602668  0.36218771]
episode: 3664000, epsilon: 0.8939110000027005, loss: 0.266993, win_rate: [ 0.2928548   0.34577585  0.36136935]
episode: 3666000, epsilon: 0.8968380000027847, loss: 0.351914, win_rate: [ 0.29381373  0.34579779  0.36038848]
episode: 3668000, epsilon: 0.8998000000028699, loss: 0.257421, win_rate: [ 0.29373097  0.34568609  0.36058293]
episode: 3670000, epsilon: 0.9027700000029553, loss: 0.319595, win_rate: [ 0.29451008  0.34525221  0.36023771]
episode: 3672000, epsilon: 0.9057290000030404, loss: 0.307719, win_rate: [ 0.29484313  0.34457855  0.36057833]
episode: 3674000, epsilon: 0.9086640000031247, loss: 0.435519, win_rate: [ 0.29553655  0.34436021  0.36010324]
episode: 3676000, epsilon: 0.9115770000032085, loss: 0.264494, win_rate: [ 0.29596979  0.34441652  0.35961369]
episode: 3678000, epsilon: 0.9145010000032926, loss: 0.199258, win_rate: [ 0.29644492  0.34420072  0.35935437]
episode: 3680000, epsilon: 0.9174530000033775, loss: 0.146098, win_rate: [ 0.29750878  0.3438832   0.35860802]
episode: 3682000, epsilon: 0.9203770000034616, loss: 0.152499, win_rate: [ 0.29802076  0.34361776  0.35836148]
episode: 3684000, epsilon: 0.9233280000035464, loss: 0.157403, win_rate: [ 0.29849645  0.34317449  0.35832907]
episode: 3686000, epsilon: 0.9262740000036311, loss: 0.431542, win_rate: [ 0.29910117  0.34263555  0.35826328]
episode: 3688000, epsilon: 0.9292130000037157, loss: 0.236954, win_rate: [ 0.29987159  0.34272338  0.35740503]
episode: 3690000, epsilon: 0.9321600000038004, loss: 0.267486, win_rate: [ 0.30075221  0.3422962   0.35695159]
episode: 3692000, epsilon: 0.9350640000038839, loss: 0.247473, win_rate: [ 0.30194237  0.3416702   0.35638743]
episode: 3694000, epsilon: 0.9379960000039682, loss: 0.268537, win_rate: [ 0.30246487  0.34137935  0.35615579]
episode: 3696000, epsilon: 0.9409320000040526, loss: 0.415605, win_rate: [ 0.30270518  0.34136103  0.35593379]
episode: 3698000, epsilon: 0.9438700000041371, loss: 0.0880194, win_rate: [ 0.30326221  0.34125162  0.35548617]
episode: 3700000, epsilon: 0.9468460000042227, loss: 0.421257, win_rate: [ 0.30419696  0.34059659  0.35520645]
episode: 3702000, epsilon: 0.9497560000043064, loss: 0.653003, win_rate: [ 0.30484015  0.34002608  0.35513377]
episode: 3704000, epsilon: 0.95, loss: 1.01045, win_rate: [ 0.30550668  0.33990058  0.35459274]
episode: 3706000, epsilon: 0.95, loss: 0.192557, win_rate: [ 0.30618579  0.33941189  0.35440232]
episode: 3708000, epsilon: 0.95, loss: 0.144085, win_rate: [ 0.30675642  0.33882094  0.35442264]
episode: 3710000, epsilon: 0.95, loss: 0.268269, win_rate: [ 0.30678812  0.33844238  0.3547695 ]
episode: 3712000, epsilon: 0.95, loss: 0.259535, win_rate: [ 0.30708654  0.33848805  0.35442541]
episode: 3714000, epsilon: 0.95, loss: 0.153976, win_rate: [ 0.30734818  0.33815493  0.35449689]
episode: 3716000, epsilon: 0.95, loss: 0.0892732, win_rate: [ 0.30785942  0.33785916  0.35428143]
episode: 3718000, epsilon: 0.95, loss: 0.208166, win_rate: [ 0.30879399  0.33731918  0.35388683]
episode: 3720000, epsilon: 0.95, loss: 0.491364, win_rate: [ 0.30898909  0.33716386  0.35384705]
episode: 3722000, epsilon: 0.95, loss: 0.153795, win_rate: [ 0.30956304  0.33670216  0.35373481]
episode: 3724000, epsilon: 0.95, loss: 0.339149, win_rate: [ 0.30955396  0.3368118   0.35363424]
episode: 3726000, epsilon: 0.95, loss: 0.304119, win_rate: [ 0.30948167  0.33675923  0.3537591 ]
episode: 3728000, epsilon: 0.95, loss: 0.149031, win_rate: [ 0.30993508  0.33645831  0.35360661]
episode: 3730000, epsilon: 0.95, loss: 0.851414, win_rate: [ 0.30984377  0.3365128   0.35364343]
episode: 3732000, epsilon: 0.95, loss: 0.101514, win_rate: [ 0.30989159  0.3365429   0.3535655 ]
episode: 3734000, epsilon: 0.95, loss: 0.39142, win_rate: [ 0.31054246  0.33618406  0.35327348]
episode: 3736000, epsilon: 0.95, loss: 0.385647, win_rate: [ 0.3108433   0.33632106  0.35283564]
episode: 3738000, epsilon: 0.95, loss: 0.563414, win_rate: [ 0.31098325  0.33618597  0.35283078]
episode: 3740000, epsilon: 0.95, loss: 0.381661, win_rate: [ 0.31131206  0.33598331  0.35270462]
episode: 3742000, epsilon: 0.95, loss: 0.204665, win_rate: [ 0.31167386  0.3354906   0.35283554]
episode: 3744000, epsilon: 0.95, loss: 0.37859, win_rate: [ 0.31184506  0.33542128  0.35273366]
episode: 3746000, epsilon: 0.95, loss: 0.187508, win_rate: [ 0.31229923  0.33514839  0.35255238]
episode: 3748000, epsilon: 0.95, loss: 0.376309, win_rate: [ 0.31255194  0.33501125  0.35243681]
episode: 3750000, epsilon: 0.95, loss: 0.322364, win_rate: [ 0.31254458  0.3350911   0.35236432]
episode: 3752000, epsilon: 0.95, loss: 0.384593, win_rate: [ 0.31264268  0.33511622  0.3522411 ]
episode: 3754000, epsilon: 0.95, loss: 0.296356, win_rate: [ 0.31282914  0.33484198  0.35232888]
episode: 3756000, epsilon: 0.95, loss: 0.312971, win_rate: [ 0.31314543  0.33459401  0.35226056]
episode: 3758000, epsilon: 0.95, loss: 0.225437, win_rate: [ 0.31338409  0.3344789   0.35213701]
episode: 3760000, epsilon: 0.95, loss: 0.330785, win_rate: [ 0.31342929  0.33429166  0.35227905]
episode: 3762000, epsilon: 0.95, loss: 0.322666, win_rate: [ 0.31359683  0.33407818  0.35232499]
episode: 3764000, epsilon: 0.95, loss: 0.109877, win_rate: [ 0.31363833  0.33413211  0.35222956]
episode: 3766000, epsilon: 0.95, loss: 0.146764, win_rate: [ 0.31412461  0.33388353  0.35199186]
episode: 3768000, epsilon: 0.95, loss: 0.254525, win_rate: [ 0.31439694  0.33351587  0.35208719]
episode: 3770000, epsilon: 0.95, loss: 0.139852, win_rate: [ 0.31456286  0.33325098  0.35218616]
episode: 3772000, epsilon: 0.95, loss: 0.136959, win_rate: [ 0.31456794  0.33316085  0.35227121]
episode: 3774000, epsilon: 0.95, loss: 0.184231, win_rate: [ 0.31490624  0.33271648  0.35237729]
episode: 3776000, epsilon: 0.95, loss: 0.410336, win_rate: [ 0.31533344  0.33248107  0.3521855 ]
episode: 3778000, epsilon: 0.95, loss: 0.263789, win_rate: [ 0.3156162   0.3323015   0.35208229]
episode: 3780000, epsilon: 0.95, loss: 0.178508, win_rate: [ 0.31574269  0.33227038  0.35198693]
episode: 3782000, epsilon: 0.95, loss: 0.419602, win_rate: [ 0.31607519  0.33188279  0.35204202]
episode: 3784000, epsilon: 0.95, loss: 0.343642, win_rate: [ 0.31655263  0.33154711  0.35190026]
episode: 3786000, epsilon: 0.95, loss: 0.186073, win_rate: [ 0.31682088  0.33163263  0.3515465 ]
episode: 3788000, epsilon: 0.95, loss: 0.689649, win_rate: [ 0.3171047   0.33146632  0.35142898]
episode: 3790000, epsilon: 0.95, loss: 0.220358, win_rate: [ 0.31745622  0.33124563  0.35129815]
episode: 3792000, epsilon: 0.95, loss: 0.206675, win_rate: [ 0.31758168  0.33118578  0.35123255]
episode: 3794000, epsilon: 0.95, loss: 0.13369, win_rate: [ 0.31807568  0.33091067  0.35101365]
episode: 3796000, epsilon: 0.95, loss: 0.160005, win_rate: [ 0.31832491  0.33079933  0.35087576]
episode: 3798000, epsilon: 0.95, loss: 0.0554987, win_rate: [ 0.31838728  0.33079126  0.35082146]
episode: 3800000, epsilon: 0.95, loss: 0.0797715, win_rate: [ 0.31839341  0.33058335  0.35102324]
episode: 3802000, epsilon: 0.95, loss: 0.0680079, win_rate: [ 0.31847367  0.33064193  0.3508844 ]
episode: 3804000, epsilon: 0.95, loss: 0.194355, win_rate: [ 0.31858177  0.33060622  0.35081201]
episode: 3806000, epsilon: 0.95, loss: 0.313856, win_rate: [ 0.31843535  0.33054208  0.35102257]
episode: 3808000, epsilon: 0.95, loss: 0.140684, win_rate: [ 0.31866193  0.33033014  0.35100793]
episode: 3810000, epsilon: 0.95, loss: 0.166439, win_rate: [ 0.31885562  0.33001271  0.35113166]
episode: 3812000, epsilon: 0.95, loss: 0.117603, win_rate: [ 0.31900793  0.32992297  0.3510691 ]
episode: 3814000, epsilon: 0.95, loss: 0.0619971, win_rate: [ 0.31920412  0.32974145  0.35105443]
episode: 3816000, epsilon: 0.95, loss: 0.905449, win_rate: [ 0.31916056  0.32971607  0.35112337]
episode: 3818000, epsilon: 0.95, loss: 0.257603, win_rate: [ 0.31938844  0.32958106  0.3510305 ]
episode: 3820000, epsilon: 0.95, loss: 0.493974, win_rate: [ 0.31955764  0.32961668  0.35082568]
episode: 3822000, epsilon: 0.95, loss: 0.42718, win_rate: [ 0.31969676  0.32958861  0.35071464]
episode: 3824000, epsilon: 0.95, loss: 0.242985, win_rate: [ 0.31977536  0.32944496  0.35077968]
episode: 3826000, epsilon: 0.95, loss: 0.515968, win_rate: [ 0.32000743  0.32941447  0.35057809]
episode: 3828000, epsilon: 0.95, loss: 0.203922, win_rate: [ 0.32006     0.32932312  0.35061688]
episode: 3830000, epsilon: 0.95, loss: 0.290547, win_rate: [ 0.32009861  0.32930292  0.35059848]
episode: 3832000, epsilon: 0.95, loss: 0.602741, win_rate: [ 0.32016672  0.32938651  0.35044677]
episode: 3834000, epsilon: 0.95, loss: 0.631074, win_rate: [ 0.3204559   0.32920372  0.35034038]
episode: 3836000, epsilon: 0.95, loss: 0.132008, win_rate: [ 0.32072745  0.32912996  0.35014258]
episode: 3838000, epsilon: 0.95, loss: 0.133709, win_rate: [ 0.32079697  0.32909105  0.35011197]
episode: 3840000, epsilon: 0.95, loss: 0.251896, win_rate: [ 0.32085283  0.32909446  0.35005271]
episode: 3842000, epsilon: 0.95, loss: 0.247613, win_rate: [ 0.32100694  0.32905649  0.34993657]
episode: 3844000, epsilon: 0.95, loss: 0.356399, win_rate: [ 0.32117901  0.32898226  0.34983873]
episode: 3846000, epsilon: 0.95, loss: 0.259129, win_rate: [ 0.32118975  0.32892143  0.34988882]
episode: 3848000, epsilon: 0.95, loss: 0.319313, win_rate: [ 0.32126887  0.32884142  0.34988972]
episode: 3850000, epsilon: 0.95, loss: 0.863928, win_rate: [ 0.32133471  0.32893068  0.3497346 ]
episode: 3852000, epsilon: 0.95, loss: 0.236154, win_rate: [ 0.32143126  0.32887965  0.34968909]
episode: 3854000, epsilon: 0.95, loss: 0.357837, win_rate: [ 0.3214948   0.32883335  0.34967185]
episode: 3856000, epsilon: 0.95, loss: 0.0710981, win_rate: [ 0.32149093  0.328862    0.34964707]
episode: 3858000, epsilon: 0.95, loss: 0.143141, win_rate: [ 0.32153751  0.32880493  0.34965756]
episode: 3860000, epsilon: 0.95, loss: 0.350192, win_rate: [ 0.32159876  0.3287872   0.34961404]
episode: 3862000, epsilon: 0.95, loss: 0.40336, win_rate: [ 0.32151022  0.32882317  0.3496666 ]
episode: 3864000, epsilon: 0.95, loss: 0.12389, win_rate: [ 0.32170711  0.32869951  0.34959337]
episode: 3866000, epsilon: 0.95, loss: 0.252253, win_rate: [ 0.32179202  0.32857395  0.34963402]
episode: 3868000, epsilon: 0.95, loss: 0.196998, win_rate: [ 0.32183462  0.32852489  0.34964049]
episode: 3870000, epsilon: 0.95, loss: 0.092305, win_rate: [ 0.32189881  0.3285136   0.34958759]
episode: 3872000, epsilon: 0.95, loss: 0.303167, win_rate: [ 0.32195838  0.32841423  0.34962739]
episode: 3874000, epsilon: 0.95, loss: 0.371829, win_rate: [ 0.32196233  0.32853895  0.34949872]
episode: 3876000, epsilon: 0.95, loss: 0.518522, win_rate: [ 0.3219155   0.32860026  0.34948424]
episode: 3878000, epsilon: 0.95, loss: 0.358542, win_rate: [ 0.32193409  0.32870745  0.34935846]
episode: 3880000, epsilon: 0.95, loss: 0.387743, win_rate: [ 0.32195242  0.32883454  0.34921304]
episode: 3882000, epsilon: 0.95, loss: 0.401359, win_rate: [ 0.32199531  0.32896337  0.34904132]
episode: 3884000, epsilon: 0.95, loss: 0.480798, win_rate: [ 0.32209042  0.32906574  0.34884384]
episode: 3886000, epsilon: 0.95, loss: 0.892699, win_rate: [ 0.32220517  0.32906179  0.34873305]
episode: 3888000, epsilon: 0.95, loss: 0.277532, win_rate: [ 0.32232527  0.32904747  0.34862726]
episode: 3890000, epsilon: 0.95, loss: 0.357506, win_rate: [ 0.32237475  0.3290299   0.34859535]
episode: 3892000, epsilon: 0.95, loss: 0.81095, win_rate: [ 0.3224989   0.32903312  0.34846798]
episode: 3894000, epsilon: 0.95, loss: 0.464274, win_rate: [ 0.32275741  0.32890704  0.34833555]
episode: 3896000, epsilon: 0.95, loss: 0.150661, win_rate: [ 0.32284013  0.32888402  0.34827585]
episode: 3898000, epsilon: 0.95, loss: 0.170224, win_rate: [ 0.32309288  0.32885796  0.34804917]
episode: 3900000, epsilon: 0.95, loss: 0.242481, win_rate: [ 0.32317892  0.32882224  0.34799884]
episode: 3902000, epsilon: 0.95, loss: 0.512195, win_rate: [ 0.32315787  0.32878368  0.34805845]
episode: 3904000, epsilon: 0.95, loss: 0.0664011, win_rate: [ 0.32324565  0.32862392  0.34813043]
episode: 3906000, epsilon: 0.95, loss: 0.0857004, win_rate: [ 0.32328326  0.32862965  0.3480871 ]
episode: 3908000, epsilon: 0.95, loss: 0.268654, win_rate: [ 0.32322947  0.32850543  0.3482651 ]
episode: 3910000, epsilon: 0.95, loss: 0.25528, win_rate: [ 0.32334734  0.32847313  0.34817952]
episode: 3912000, epsilon: 0.95, loss: 0.181956, win_rate: [ 0.32343486  0.32836113  0.34820401]
episode: 3914000, epsilon: 0.95, loss: 0.063762, win_rate: [ 0.32358814  0.32831424  0.34809762]
episode: 3916000, epsilon: 0.95, loss: 0.365352, win_rate: [ 0.3236129  0.3282458  0.3481413]
episode: 3918000, epsilon: 0.95, loss: 0.384387, win_rate: [ 0.32363735  0.32837318  0.34798947]
episode: 3920000, epsilon: 0.95, loss: 0.487182, win_rate: [ 0.32369586  0.32839897  0.34790516]
episode: 3922000, epsilon: 0.95, loss: 0.637752, win_rate: [ 0.32363875  0.32850209  0.34785917]
episode: 3924000, epsilon: 0.95, loss: 0.294062, win_rate: [ 0.32366567  0.32859775  0.34773658]
episode: 3926000, epsilon: 0.95, loss: 0.619658, win_rate: [ 0.3236125   0.32867384  0.34771366]
episode: 3928000, epsilon: 0.95, loss: 0.529847, win_rate: [ 0.32346548  0.32872766  0.34780687]
episode: 3930000, epsilon: 0.95, loss: 0.150985, win_rate: [ 0.32360508  0.32864446  0.34775046]
episode: 3932000, epsilon: 0.95, loss: 0.972877, win_rate: [ 0.3236436   0.32878515  0.34757124]
episode: 3934000, epsilon: 0.95, loss: 4.92489, win_rate: [ 0.32354095  0.32885231  0.34760674]
episode: 3936000, epsilon: 0.95, loss: 1.1805, win_rate: [ 0.32360023  0.32882343  0.34757635]
episode: 3938000, epsilon: 0.95, loss: 2.09777, win_rate: [ 0.32362626  0.32893394  0.3474398 ]
episode: 3940000, epsilon: 0.8002380000000069, loss: 2.42326, win_rate: [ 0.32347258  0.32919903  0.34732839]
episode: 3942000, epsilon: 0.8032210000000927, loss: 0.60231, win_rate: [ 0.32306923  0.32937623  0.34755454]
episode: 3944000, epsilon: 0.8061970000001782, loss: 1.85586, win_rate: [ 0.32265022  0.32965311  0.34769666]
episode: 3946000, epsilon: 0.8091870000002642, loss: 0.329816, win_rate: [ 0.32226785  0.32984587  0.34788628]
episode: 3948000, epsilon: 0.8121750000003501, loss: 0.467693, win_rate: [ 0.32182379  0.33004503  0.34813118]
episode: 3950000, epsilon: 0.8151770000004365, loss: 0.324925, win_rate: [ 0.32147051  0.3302162   0.34831329]
episode: 3952000, epsilon: 0.8181690000005225, loss: 0.526728, win_rate: [ 0.32113261  0.33039963  0.34846776]
episode: 3954000, epsilon: 0.8211330000006077, loss: 0.478741, win_rate: [ 0.32074203  0.33058099  0.34867698]
episode: 3956000, epsilon: 0.8240740000006923, loss: 1.16569, win_rate: [ 0.32044292  0.33082772  0.34872936]
episode: 3958000, epsilon: 0.8270290000007773, loss: 1.37704, win_rate: [ 0.32007173  0.33106332  0.34886495]
episode: 3960000, epsilon: 0.8299930000008625, loss: 0.664318, win_rate: [ 0.319788    0.33124908  0.34896292]
episode: 3962000, epsilon: 0.8329450000009474, loss: 0.320014, win_rate: [ 0.31958751  0.33139411  0.34901837]
episode: 3964000, epsilon: 0.8358920000010321, loss: 0.401112, win_rate: [ 0.31936725  0.33155953  0.34907322]
episode: 3966000, epsilon: 0.8388200000011163, loss: 1.74658, win_rate: [ 0.31900186  0.3318051   0.34919303]
episode: 3968000, epsilon: 0.8417850000012016, loss: 0.699937, win_rate: [ 0.31871109  0.33204529  0.34924362]
episode: 3970000, epsilon: 0.8447440000012867, loss: 0.651783, win_rate: [ 0.31853157  0.33228829  0.34918014]
episode: 3972000, epsilon: 0.8477020000013717, loss: 0.6139, win_rate: [ 0.3185099   0.33240771  0.34908239]
episode: 3974000, epsilon: 0.8506570000014567, loss: 0.5426, win_rate: [ 0.31844835  0.33243761  0.34911404]
episode: 3976000, epsilon: 0.8536100000015416, loss: 0.836966, win_rate: [ 0.31832362  0.3326055   0.34907088]
episode: 3978000, epsilon: 0.8565670000016267, loss: 0.8269, win_rate: [ 0.31823196  0.33268695  0.34908109]
episode: 3980000, epsilon: 0.8594960000017109, loss: 0.861648, win_rate: [ 0.31820443  0.33276491  0.34903066]
episode: 3982000, epsilon: 0.8624490000017958, loss: 0.641001, win_rate: [ 0.31804891  0.33288656  0.34906453]
episode: 3984000, epsilon: 0.8654030000018808, loss: 1.1885, win_rate: [ 0.31794448  0.33296267  0.34909284]
episode: 3986000, epsilon: 0.8683650000019659, loss: 0.247038, win_rate: [ 0.3178256  0.332981   0.3491934]
episode: 3988000, epsilon: 0.8712980000020503, loss: 0.358208, win_rate: [ 0.31768475  0.3329012   0.34941405]
episode: 3990000, epsilon: 0.8742540000021353, loss: 0.338295, win_rate: [ 0.31775047  0.33285043  0.3493991 ]
episode: 3992000, epsilon: 0.8772120000022203, loss: 0.442767, win_rate: [ 0.31758592  0.33287415  0.34953992]
episode: 3994000, epsilon: 0.8801530000023049, loss: 0.257307, win_rate: [ 0.31750935  0.33286718  0.34962348]
episode: 3996000, epsilon: 0.8831100000023899, loss: 0.201078, win_rate: [ 0.31738304  0.33285017  0.34976679]
episode: 3998000, epsilon: 0.886070000002475, loss: 0.34941, win_rate: [ 0.31721528  0.33290117  0.34988354]
episode: 4000000, epsilon: 0.8890290000025601, loss: 0.0794598, win_rate: [ 0.31696421  0.33290917  0.35012662]
episode: 4002000, epsilon: 0.8919600000026444, loss: 0.494786, win_rate: [ 0.31695444  0.3329345   0.35011107]
episode: 4004000, epsilon: 0.8949260000027297, loss: 0.0423831, win_rate: [ 0.31698436  0.33292244  0.35009319]
episode: 0, win_rate: [ 0.  0.  1.]
episode: 0, win_rate: [ 0.  0.  1.]
episode: 0, win_rate: [ 0.  0.  1.]
episode: 0, win_rate: [ 0.  0.  1.]
episode: 0, win_rate: [ 0.  0.  1.]
episode: 0, win_rate: [ 1.  0.  0.]
episode: 0, win_rate: [ 0.  0.  1.]
episode: 0, win_rate: [ 0.  0.  1.]
episode: 0, win_rate: [ 1.  0.  0.]
episode: 0, win_rate: [ 0.  1.  0.]
episode: 2000, win_rate: [ 0.33133433  0.33733133  0.33133433]
episode: 0, win_rate: [ 1.  0.  0.]
episode: 2000, win_rate: [ 0.34632684  0.32233883  0.33133433]
episode: 4000, win_rate: [ 0.3484129   0.32691827  0.32466883]
episode: 6000, win_rate: [ 0.34410932  0.32894518  0.32694551]
episode: 8000, win_rate: [ 0.34808149  0.32595926  0.32595926]
episode: 0, win_rate: [ 0.  1.  0.]
episode: 2000, win_rate: [ 0.34582709  0.32233883  0.33183408]
episode: 4000, win_rate: [ 0.33316671  0.33366658  0.33316671]
episode: 0, win_rate: [ 0.  1.  0.]
episode: 2000, win_rate: [ 0.34282859  0.32833583  0.32883558]
episode: 4000, win_rate: [ 0.34716321  0.32116971  0.33166708]
episode: 6000, win_rate: [ 0.34427595  0.32761206  0.32811198]
episode: 8000, win_rate: [ 0.34383202  0.32733408  0.3288339 ]
episode: 0, win_rate: [ 1.  0.  0.]
episode: 0, win_rate: [ 1.  0.  0.]
episode: 0, win_rate: [ 1.  0.  0.]
episode: 0, win_rate: [ 1.  0.  0.]
episode: 0, win_rate: [ 1.  0.  0.]
episode: 0, win_rate: [ 1.  0.  0.]
episode: 0, win_rate: [ 0.  0.  1.]
episode: 0, win_rate: [ 0.  1.  0.]
episode: 0, win_rate: [ 0.  0.  1.]
episode: 0, win_rate: [ 1.  0.  0.]
episode: 0, win_rate: [ 0.  0.  1.]
episode: 0, win_rate: [ 0.  1.  0.]
episode: 0, win_rate: [ 0.  1.  0.]
episode: 0, win_rate: [ 0.  0.  1.]
episode: 0, win_rate: [ 0.  1.  0.]
episode: 0, win_rate: [ 1.  0.  0.]
episode: 0, win_rate: [ 0.  1.  0.]
episode: 0, win_rate: [ 0.  0.  1.]
episode: 0, win_rate: [ 1.  0.  0.]
episode: 0, win_rate: [ 0.  1.  0.]
episode: 0, win_rate: [ 1.  0.  0.]
episode: 0, win_rate: [ 1.  0.  0.]
episode: 0, win_rate: [ 1.  0.  0.]
episode: 0, win_rate: [ 0.  1.  0.]
episode: 0, win_rate: [ 1.  0.  0.]
episode: 0, win_rate: [ 0.  0.  1.]
episode: 0, win_rate: [ 0.  0.  1.]
episode: 0, win_rate: [ 1.  0.  0.]
episode: 0, win_rate: [ 0.  0.  1.]
episode: 0, win_rate: [ 0.  1.  0.]
episode: 0, win_rate: [ 1.  0.  0.]
episode: 0, win_rate: [ 0.  0.  1.]
episode: 0, win_rate: [ 0.  0.  1.]
episode: 0, win_rate: [ 1.  0.  0.]
episode: 0, win_rate: [ 0.  1.  0.]
episode: 0, win_rate: [ 1.  0.  0.]
episode: 0, win_rate: [ 1.  0.  0.]
episode: 0, win_rate: [ 1.  0.  0.]
episode: 0, win_rate: [ 1.  0.  0.]
episode: 0, win_rate: [ 0.  1.  0.]