agent$learn(500)
## Episode: 1 finished with steps:55, rewards:-55.000000 global step 55
## Last 100 episodes average reward -55.000000
## Epsilon0.099750
## rand steps:6
## Episode: 2 finished with steps:632, rewards:-2414.000000 global step 687
## Last 100 episodes average reward -1234.500000
## Epsilon0.099500
## rand steps:72
## Episode: 3 finished with steps:36, rewards:-36.000000 global step 723
## Last 100 episodes average reward -835.000000
## Epsilon0.099250
## rand steps:2
## Episode: 4 finished with steps:143, rewards:-440.000000 global step 866
## Last 100 episodes average reward -736.250000
## Epsilon0.099000
## rand steps:19
## Episode: 5 finished with steps:85, rewards:-85.000000 global step 951
## Last 100 episodes average reward -606.000000
## Epsilon0.098750
## rand steps:7
## Episode: 6 finished with steps:54, rewards:-54.000000 global step 1005
## Last 100 episodes average reward -514.000000
## Epsilon0.098500
## rand steps:6
## Episode: 7 finished with steps:143, rewards:-341.000000 global step 1148
## Last 100 episodes average reward -489.285714
## Epsilon0.098250
## rand steps:18
## Episode: 8 finished with steps:58, rewards:-58.000000 global step 1206
## Last 100 episodes average reward -435.375000
## Epsilon0.098000
## rand steps:8
## Episode: 9 finished with steps:95, rewards:-194.000000 global step 1301
## Last 100 episodes average reward -408.555556
## Epsilon0.097750
## rand steps:11
## Episode: 10 finished with steps:68, rewards:-167.000000 global step 1369
## Last 100 episodes average reward -384.400000
## Epsilon0.097500
## rand steps:6
## Episode: 11 finished with steps:51, rewards:-51.000000 global step 1420
## Last 100 episodes average reward -354.090909
## Epsilon0.097250
## rand steps:4
## Episode: 12 finished with steps:35, rewards:-35.000000 global step 1455
## Last 100 episodes average reward -327.500000
## Epsilon0.097000
## rand steps:4
## Episode: 13 finished with steps:45, rewards:-45.000000 global step 1500
## Last 100 episodes average reward -305.769231
## Epsilon0.096750
## rand steps:3
## Episode: 14 finished with steps:30, rewards:-30.000000 global step 1530
## Last 100 episodes average reward -286.071429
## Epsilon0.096500
## rand steps:4
## Episode: 15 finished with steps:68, rewards:-68.000000 global step 1598
## Last 100 episodes average reward -271.533333
## Epsilon0.096250
## rand steps:7
## Episode: 16 finished with steps:26, rewards:-26.000000 global step 1624
## Last 100 episodes average reward -256.187500
## Epsilon0.096000
## rand steps:3
## Episode: 17 finished with steps:33, rewards:-33.000000 global step 1657
## Last 100 episodes average reward -243.058824
## Epsilon0.095750
## rand steps:3
## Episode: 18 finished with steps:53, rewards:-53.000000 global step 1710
## Last 100 episodes average reward -232.500000
## Epsilon0.095500
## rand steps:5
## Episode: 19 finished with steps:27, rewards:-27.000000 global step 1737
## Last 100 episodes average reward -221.684211
## Epsilon0.095250
## rand steps:1
## Episode: 20 finished with steps:19, rewards:-19.000000 global step 1756
## Last 100 episodes average reward -211.550000
## Epsilon0.095000
## rand steps:0
## Episode: 21 finished with steps:56, rewards:-56.000000 global step 1812
## Last 100 episodes average reward -204.142857
## Epsilon0.094750
## rand steps:4
## Episode: 22 finished with steps:41, rewards:-41.000000 global step 1853
## Last 100 episodes average reward -196.727273
## Epsilon0.094500
## rand steps:0
## Episode: 23 finished with steps:25, rewards:-25.000000 global step 1878
## Last 100 episodes average reward -189.260870
## Epsilon0.094250
## rand steps:5
## Episode: 24 finished with steps:21, rewards:-21.000000 global step 1899
## Last 100 episodes average reward -182.250000
## Epsilon0.094000
## rand steps:4
## Episode: 25 finished with steps:42, rewards:-42.000000 global step 1941
## Last 100 episodes average reward -176.640000
## Epsilon0.093750
## rand steps:2
## Episode: 26 finished with steps:34, rewards:-34.000000 global step 1975
## Last 100 episodes average reward -171.153846
## Epsilon0.093500
## rand steps:3
## Episode: 27 finished with steps:49, rewards:-49.000000 global step 2024
## Last 100 episodes average reward -166.629630
## Epsilon0.093250
## rand steps:4
## Episode: 28 finished with steps:63, rewards:-360.000000 global step 2087
## Last 100 episodes average reward -173.535714
## Epsilon0.093000
## rand steps:8
## Episode: 29 finished with steps:29, rewards:-29.000000 global step 2116
## Last 100 episodes average reward -168.551724
## Epsilon0.092750
## rand steps:4
## Episode: 30 finished with steps:29, rewards:-29.000000 global step 2145
## Last 100 episodes average reward -163.900000
## Epsilon0.092500
## rand steps:3
## Episode: 31 finished with steps:23, rewards:-23.000000 global step 2168
## Last 100 episodes average reward -159.354839
## Epsilon0.092250
## rand steps:2
## Episode: 32 finished with steps:20, rewards:-20.000000 global step 2188
## Last 100 episodes average reward -155.000000
## Epsilon0.092000
## rand steps:2
## Episode: 33 finished with steps:23, rewards:-23.000000 global step 2211
## Last 100 episodes average reward -151.000000
## Epsilon0.091750
## rand steps:0
## Episode: 34 finished with steps:25, rewards:-25.000000 global step 2236
## Last 100 episodes average reward -147.294118
## Epsilon0.091500
## rand steps:1
## Episode: 35 finished with steps:26, rewards:-26.000000 global step 2262
## Last 100 episodes average reward -143.828571
## Epsilon0.091250
## rand steps:4
## Episode: 36 finished with steps:27, rewards:-27.000000 global step 2289
## Last 100 episodes average reward -140.583333
## Epsilon0.091000
## rand steps:3
## Episode: 37 finished with steps:26, rewards:-125.000000 global step 2315
## Last 100 episodes average reward -140.162162
## Epsilon0.090750
## rand steps:2
## Episode: 38 finished with steps:21, rewards:-21.000000 global step 2336
## Last 100 episodes average reward -137.026316
## Epsilon0.090500
## rand steps:2
## Episode: 39 finished with steps:44, rewards:-44.000000 global step 2380
## Last 100 episodes average reward -134.641026
## Epsilon0.090250
## rand steps:4
## Episode: 40 finished with steps:17, rewards:-17.000000 global step 2397
## Last 100 episodes average reward -131.700000
## Epsilon0.090000
## rand steps:2
## Episode: 41 finished with steps:22, rewards:-22.000000 global step 2419
## Last 100 episodes average reward -129.024390
## Epsilon0.089750
## rand steps:1
## Episode: 42 finished with steps:30, rewards:-30.000000 global step 2449
## Last 100 episodes average reward -126.666667
## Epsilon0.089500
## rand steps:3
## Episode: 43 finished with steps:26, rewards:-26.000000 global step 2475
## Last 100 episodes average reward -124.325581
## Epsilon0.089250
## rand steps:1
## Episode: 44 finished with steps:22, rewards:-22.000000 global step 2497
## Last 100 episodes average reward -122.000000
## Epsilon0.089000
## rand steps:3
## Episode: 45 finished with steps:13, rewards:-13.000000 global step 2510
## Last 100 episodes average reward -119.577778
## Epsilon0.088750
## rand steps:1
## Episode: 46 finished with steps:34, rewards:-232.000000 global step 2544
## Last 100 episodes average reward -122.021739
## Epsilon0.088500
## rand steps:2
## Episode: 47 finished with steps:15, rewards:-15.000000 global step 2559
## Last 100 episodes average reward -119.744681
## Epsilon0.088250
## rand steps:0
## Episode: 48 finished with steps:14, rewards:-14.000000 global step 2573
## Last 100 episodes average reward -117.541667
## Epsilon0.088000
## rand steps:0
## Episode: 49 finished with steps:23, rewards:-122.000000 global step 2596
## Last 100 episodes average reward -117.632653
## Epsilon0.087750
## rand steps:1
## Episode: 50 finished with steps:24, rewards:-24.000000 global step 2620
## Last 100 episodes average reward -115.760000
## Epsilon0.087500
## rand steps:1
## Episode: 51 finished with steps:29, rewards:-128.000000 global step 2649
## Last 100 episodes average reward -116.000000
## Epsilon0.087250
## rand steps:4
## Episode: 52 finished with steps:19, rewards:-19.000000 global step 2668
## Last 100 episodes average reward -114.134615
## Epsilon0.087000
## rand steps:2
## Episode: 53 finished with steps:32, rewards:-32.000000 global step 2700
## Last 100 episodes average reward -112.584906
## Epsilon0.086750
## rand steps:6
## Episode: 54 finished with steps:13, rewards:-13.000000 global step 2713
## Last 100 episodes average reward -110.740741
## Epsilon0.086500
## rand steps:0
## Episode: 55 finished with steps:20, rewards:-119.000000 global step 2733
## Last 100 episodes average reward -110.890909
## Epsilon0.086250
## rand steps:1
## Episode: 56 finished with steps:17, rewards:-17.000000 global step 2750
## Last 100 episodes average reward -109.214286
## Epsilon0.086000
## rand steps:3
## Episode: 57 finished with steps:15, rewards:-15.000000 global step 2765
## Last 100 episodes average reward -107.561404
## Epsilon0.085750
## rand steps:2
## Episode: 58 finished with steps:14, rewards:-14.000000 global step 2779
## Last 100 episodes average reward -105.948276
## Epsilon0.085500
## rand steps:1
## Episode: 59 finished with steps:13, rewards:-13.000000 global step 2792
## Last 100 episodes average reward -104.372881
## Epsilon0.085250
## rand steps:0
## Episode: 60 finished with steps:34, rewards:-133.000000 global step 2826
## Last 100 episodes average reward -104.850000
## Epsilon0.085000
## rand steps:7
## Episode: 61 finished with steps:13, rewards:-13.000000 global step 2839
## Last 100 episodes average reward -103.344262
## Epsilon0.084750
## rand steps:0
## Episode: 62 finished with steps:13, rewards:-13.000000 global step 2852
## Last 100 episodes average reward -101.887097
## Epsilon0.084500
## rand steps:0
## Episode: 63 finished with steps:22, rewards:-220.000000 global step 2874
## Last 100 episodes average reward -103.761905
## Epsilon0.084250
## rand steps:2
## Episode: 64 finished with steps:13, rewards:-13.000000 global step 2887
## Last 100 episodes average reward -102.343750
## Epsilon0.084000
## rand steps:0
## Episode: 65 finished with steps:16, rewards:-115.000000 global step 2903
## Last 100 episodes average reward -102.538462
## Epsilon0.083750
## rand steps:1
## Episode: 66 finished with steps:13, rewards:-13.000000 global step 2916
## Last 100 episodes average reward -101.181818
## Epsilon0.083500
## rand steps:0
## Episode: 67 finished with steps:19, rewards:-118.000000 global step 2935
## Last 100 episodes average reward -101.432836
## Epsilon0.083250
## rand steps:2
## Episode: 68 finished with steps:13, rewards:-13.000000 global step 2948
## Last 100 episodes average reward -100.132353
## Epsilon0.083000
## rand steps:0
## Episode: 69 finished with steps:13, rewards:-13.000000 global step 2961
## Last 100 episodes average reward -98.869565
## Epsilon0.082750
## rand steps:0
## Episode: 70 finished with steps:20, rewards:-119.000000 global step 2981
## Last 100 episodes average reward -99.157143
## Epsilon0.082500
## rand steps:3
## Episode: 71 finished with steps:15, rewards:-15.000000 global step 2996
## Last 100 episodes average reward -97.971831
## Epsilon0.082250
## rand steps:1
## Episode: 72 finished with steps:35, rewards:-134.000000 global step 3031
## Last 100 episodes average reward -98.472222
## Epsilon0.082000
## rand steps:4
## Episode: 73 finished with steps:22, rewards:-121.000000 global step 3053
## Last 100 episodes average reward -98.780822
## Epsilon0.081750
## rand steps:2
## Episode: 74 finished with steps:29, rewards:-227.000000 global step 3082
## Last 100 episodes average reward -100.513514
## Epsilon0.081500
## rand steps:4
## Episode: 75 finished with steps:15, rewards:-15.000000 global step 3097
## Last 100 episodes average reward -99.373333
## Epsilon0.081250
## rand steps:1
## Episode: 76 finished with steps:13, rewards:-13.000000 global step 3110
## Last 100 episodes average reward -98.236842
## Epsilon0.081000
## rand steps:0
## Episode: 77 finished with steps:13, rewards:-13.000000 global step 3123
## Last 100 episodes average reward -97.129870
## Epsilon0.080750
## rand steps:2
## Episode: 78 finished with steps:13, rewards:-13.000000 global step 3136
## Last 100 episodes average reward -96.051282
## Epsilon0.080500
## rand steps:1
## Episode: 79 finished with steps:15, rewards:-15.000000 global step 3151
## Last 100 episodes average reward -95.025316
## Epsilon0.080250
## rand steps:1
## Episode: 80 finished with steps:13, rewards:-13.000000 global step 3164
## Last 100 episodes average reward -94.000000
## Epsilon0.080000
## rand steps:0
## Episode: 81 finished with steps:13, rewards:-13.000000 global step 3177
## Last 100 episodes average reward -93.000000
## Epsilon0.079750
## rand steps:0
## Episode: 82 finished with steps:22, rewards:-121.000000 global step 3199
## Last 100 episodes average reward -93.341463
## Epsilon0.079500
## rand steps:3
## Episode: 83 finished with steps:15, rewards:-15.000000 global step 3214
## Last 100 episodes average reward -92.397590
## Epsilon0.079250
## rand steps:1
## Episode: 84 finished with steps:13, rewards:-13.000000 global step 3227
## Last 100 episodes average reward -91.452381
## Epsilon0.079000
## rand steps:0
## Episode: 85 finished with steps:17, rewards:-17.000000 global step 3244
## Last 100 episodes average reward -90.576471
## Epsilon0.078750
## rand steps:4
## Episode: 86 finished with steps:13, rewards:-13.000000 global step 3257
## Last 100 episodes average reward -89.674419
## Epsilon0.078500
## rand steps:0
## Episode: 87 finished with steps:17, rewards:-17.000000 global step 3274
## Last 100 episodes average reward -88.839080
## Epsilon0.078250
## rand steps:3
## Episode: 88 finished with steps:16, rewards:-16.000000 global step 3290
## Last 100 episodes average reward -88.011364
## Epsilon0.078000
## rand steps:2
## Episode: 89 finished with steps:31, rewards:-229.000000 global step 3321
## Last 100 episodes average reward -89.595506
## Epsilon0.077750
## rand steps:3
## Episode: 90 finished with steps:15, rewards:-15.000000 global step 3336
## Last 100 episodes average reward -88.766667
## Epsilon0.077500
## rand steps:1
## Episode: 91 finished with steps:15, rewards:-15.000000 global step 3351
## Last 100 episodes average reward -87.956044
## Epsilon0.077250
## rand steps:1
## Episode: 92 finished with steps:13, rewards:-13.000000 global step 3364
## Last 100 episodes average reward -87.141304
## Epsilon0.077000
## rand steps:0
## Episode: 93 finished with steps:13, rewards:-13.000000 global step 3377
## Last 100 episodes average reward -86.344086
## Epsilon0.076750
## rand steps:1
## Episode: 94 finished with steps:17, rewards:-17.000000 global step 3394
## Last 100 episodes average reward -85.606383
## Epsilon0.076500
## rand steps:2
## Episode: 95 finished with steps:13, rewards:-13.000000 global step 3407
## Last 100 episodes average reward -84.842105
## Epsilon0.076250
## rand steps:0
## Episode: 96 finished with steps:13, rewards:-13.000000 global step 3420
## Last 100 episodes average reward -84.093750
## Epsilon0.076000
## rand steps:1
## Episode: 97 finished with steps:24, rewards:-123.000000 global step 3444
## Last 100 episodes average reward -84.494845
## Epsilon0.075750
## rand steps:2
## Episode: 98 finished with steps:15, rewards:-15.000000 global step 3459
## Last 100 episodes average reward -83.785714
## Epsilon0.075500
## rand steps:1
## Episode: 99 finished with steps:31, rewards:-229.000000 global step 3490
## Last 100 episodes average reward -85.252525
## Epsilon0.075250
## rand steps:6
## Episode: 100 finished with steps:19, rewards:-118.000000 global step 3509
## Last 100 episodes average reward -85.580000
## Epsilon0.075000
## rand steps:2
## Episode: 101 finished with steps:24, rewards:-123.000000 global step 3533
## Last 100 episodes average reward -85.950495
## Epsilon0.074750
## rand steps:2
## Episode: 102 finished with steps:22, rewards:-121.000000 global step 3555
## Last 100 episodes average reward -86.603960
## Epsilon0.074500
## rand steps:4
## Episode: 103 finished with steps:15, rewards:-15.000000 global step 3570
## Last 100 episodes average reward -62.851485
## Epsilon0.074250
## rand steps:2
## Episode: 104 finished with steps:23, rewards:-122.000000 global step 3593
## Last 100 episodes average reward -63.702970
## Epsilon0.074000
## rand steps:3
## Episode: 105 finished with steps:13, rewards:-13.000000 global step 3606
## Last 100 episodes average reward -59.475248
## Epsilon0.073750
## rand steps:0
## Episode: 106 finished with steps:15, rewards:-15.000000 global step 3621
## Last 100 episodes average reward -58.782178
## Epsilon0.073500
## rand steps:1
## Episode: 107 finished with steps:13, rewards:-13.000000 global step 3634
## Last 100 episodes average reward -58.376238
## Epsilon0.073250
## rand steps:0
## Episode: 108 finished with steps:13, rewards:-13.000000 global step 3647
## Last 100 episodes average reward -55.128713
## Epsilon0.073000
## rand steps:1
## Episode: 109 finished with steps:28, rewards:-127.000000 global step 3675
## Last 100 episodes average reward -55.811881
## Epsilon0.072750
## rand steps:4
## Episode: 110 finished with steps:13, rewards:-13.000000 global step 3688
## Last 100 episodes average reward -54.019802
## Epsilon0.072500
## rand steps:0
## Episode: 111 finished with steps:13, rewards:-13.000000 global step 3701
## Last 100 episodes average reward -52.495050
## Epsilon0.072250
## rand steps:0
## Episode: 112 finished with steps:14, rewards:-14.000000 global step 3715
## Last 100 episodes average reward -52.128713
## Epsilon0.072000
## rand steps:1
## Episode: 113 finished with steps:16, rewards:-115.000000 global step 3731
## Last 100 episodes average reward -52.920792
## Epsilon0.071750
## rand steps:2
## Episode: 114 finished with steps:14, rewards:-14.000000 global step 3745
## Last 100 episodes average reward -52.613861
## Epsilon0.071500
## rand steps:1
## Episode: 115 finished with steps:17, rewards:-116.000000 global step 3762
## Last 100 episodes average reward -53.465347
## Epsilon0.071250
## rand steps:3
## Episode: 116 finished with steps:24, rewards:-222.000000 global step 3786
## Last 100 episodes average reward -54.990099
## Epsilon0.071000
## rand steps:3
## Episode: 117 finished with steps:13, rewards:-13.000000 global step 3799
## Last 100 episodes average reward -54.861386
## Epsilon0.070750
## rand steps:0
## Episode: 118 finished with steps:13, rewards:-13.000000 global step 3812
## Last 100 episodes average reward -54.663366
## Epsilon0.070500
## rand steps:0
## Episode: 119 finished with steps:13, rewards:-13.000000 global step 3825
## Last 100 episodes average reward -54.267327
## Epsilon0.070250
## rand steps:0
## Episode: 120 finished with steps:13, rewards:-13.000000 global step 3838
## Last 100 episodes average reward -54.128713
## Epsilon0.070000
## rand steps:0
## Episode: 121 finished with steps:15, rewards:-15.000000 global step 3853
## Last 100 episodes average reward -54.089109
## Epsilon0.069750
## rand steps:2
## Episode: 122 finished with steps:13, rewards:-13.000000 global step 3866
## Last 100 episodes average reward -53.663366
## Epsilon0.069500
## rand steps:0
## Episode: 123 finished with steps:17, rewards:-17.000000 global step 3883
## Last 100 episodes average reward -53.425743
## Epsilon0.069250
## rand steps:2
## Episode: 124 finished with steps:13, rewards:-13.000000 global step 3896
## Last 100 episodes average reward -53.306931
## Epsilon0.069000
## rand steps:1
## Episode: 125 finished with steps:20, rewards:-20.000000 global step 3916
## Last 100 episodes average reward -53.297030
## Epsilon0.068750
## rand steps:3
## Episode: 126 finished with steps:13, rewards:-13.000000 global step 3929
## Last 100 episodes average reward -53.009901
## Epsilon0.068500
## rand steps:0
## Episode: 127 finished with steps:26, rewards:-26.000000 global step 3955
## Last 100 episodes average reward -52.930693
## Epsilon0.068250
## rand steps:2
## Episode: 128 finished with steps:23, rewards:-122.000000 global step 3978
## Last 100 episodes average reward -53.653465
## Epsilon0.068000
## rand steps:1
## Episode: 129 finished with steps:13, rewards:-13.000000 global step 3991
## Last 100 episodes average reward -50.217822
## Epsilon0.067750
## rand steps:1
## Episode: 130 finished with steps:13, rewards:-13.000000 global step 4004
## Last 100 episodes average reward -50.059406
## Epsilon0.067500
## rand steps:0
## Episode: 131 finished with steps:13, rewards:-13.000000 global step 4017
## Last 100 episodes average reward -49.900990
## Epsilon0.067250
## rand steps:1
## Episode: 132 finished with steps:13, rewards:-13.000000 global step 4030
## Last 100 episodes average reward -49.801980
## Epsilon0.067000
## rand steps:0
## Episode: 133 finished with steps:13, rewards:-13.000000 global step 4043
## Last 100 episodes average reward -49.732673
## Epsilon0.066750
## rand steps:0
## Episode: 134 finished with steps:13, rewards:-13.000000 global step 4056
## Last 100 episodes average reward -49.633663
## Epsilon0.066500
## rand steps:0
## Episode: 135 finished with steps:13, rewards:-13.000000 global step 4069
## Last 100 episodes average reward -49.514851
## Epsilon0.066250
## rand steps:0
## Episode: 136 finished with steps:13, rewards:-13.000000 global step 4082
## Last 100 episodes average reward -49.386139
## Epsilon0.066000
## rand steps:0
## Episode: 137 finished with steps:13, rewards:-13.000000 global step 4095
## Last 100 episodes average reward -49.247525
## Epsilon0.065750
## rand steps:0
## Episode: 138 finished with steps:13, rewards:-13.000000 global step 4108
## Last 100 episodes average reward -48.138614
## Epsilon0.065500
## rand steps:0
## Episode: 139 finished with steps:23, rewards:-122.000000 global step 4131
## Last 100 episodes average reward -49.138614
## Epsilon0.065250
## rand steps:4
## Episode: 140 finished with steps:13, rewards:-13.000000 global step 4144
## Last 100 episodes average reward -48.831683
## Epsilon0.065000
## rand steps:0
## Episode: 141 finished with steps:17, rewards:-116.000000 global step 4161
## Last 100 episodes average reward -49.811881
## Epsilon0.064750
## rand steps:1
## Episode: 142 finished with steps:13, rewards:-13.000000 global step 4174
## Last 100 episodes average reward -49.722772
## Epsilon0.064500
## rand steps:0
## Episode: 143 finished with steps:13, rewards:-13.000000 global step 4187
## Last 100 episodes average reward -49.554455
## Epsilon0.064250
## rand steps:0
## Episode: 144 finished with steps:17, rewards:-17.000000 global step 4204
## Last 100 episodes average reward -49.465347
## Epsilon0.064000
## rand steps:3
## Episode: 145 finished with steps:13, rewards:-13.000000 global step 4217
## Last 100 episodes average reward -49.376238
## Epsilon0.063750
## rand steps:0
## Episode: 146 finished with steps:15, rewards:-15.000000 global step 4232
## Last 100 episodes average reward -49.396040
## Epsilon0.063500
## rand steps:2
## Episode: 147 finished with steps:13, rewards:-13.000000 global step 4245
## Last 100 episodes average reward -47.227723
## Epsilon0.063250
## rand steps:0
## Episode: 148 finished with steps:13, rewards:-13.000000 global step 4258
## Last 100 episodes average reward -47.207921
## Epsilon0.063000
## rand steps:1
## Episode: 149 finished with steps:13, rewards:-13.000000 global step 4271
## Last 100 episodes average reward -47.198020
## Epsilon0.062750
## rand steps:0
## Episode: 150 finished with steps:18, rewards:-18.000000 global step 4289
## Last 100 episodes average reward -46.168317
## Epsilon0.062500
## rand steps:4
## Episode: 151 finished with steps:13, rewards:-13.000000 global step 4302
## Last 100 episodes average reward -46.059406
## Epsilon0.062250
## rand steps:0
## Episode: 152 finished with steps:13, rewards:-13.000000 global step 4315
## Last 100 episodes average reward -44.920792
## Epsilon0.062000
## rand steps:0
## Episode: 153 finished with steps:15, rewards:-15.000000 global step 4330
## Last 100 episodes average reward -44.881188
## Epsilon0.061750
## rand steps:1
## Episode: 154 finished with steps:13, rewards:-13.000000 global step 4343
## Last 100 episodes average reward -44.693069
## Epsilon0.061500
## rand steps:0
## Episode: 155 finished with steps:15, rewards:-15.000000 global step 4358
## Last 100 episodes average reward -44.712871
## Epsilon0.061250
## rand steps:2
## Episode: 156 finished with steps:13, rewards:-13.000000 global step 4371
## Last 100 episodes average reward -43.663366
## Epsilon0.061000
## rand steps:0
## Episode: 157 finished with steps:13, rewards:-13.000000 global step 4384
## Last 100 episodes average reward -43.623762
## Epsilon0.060750
## rand steps:2
## Episode: 158 finished with steps:13, rewards:-13.000000 global step 4397
## Last 100 episodes average reward -43.603960
## Epsilon0.060500
## rand steps:0
## Episode: 159 finished with steps:15, rewards:-15.000000 global step 4412
## Last 100 episodes average reward -43.613861
## Epsilon0.060250
## rand steps:1
## Episode: 160 finished with steps:13, rewards:-13.000000 global step 4425
## Last 100 episodes average reward -43.613861
## Epsilon0.060000
## rand steps:0
## Episode: 161 finished with steps:17, rewards:-17.000000 global step 4442
## Last 100 episodes average reward -42.465347
## Epsilon0.059750
## rand steps:2
## Episode: 162 finished with steps:17, rewards:-215.000000 global step 4459
## Last 100 episodes average reward -44.465347
## Epsilon0.059500
## rand steps:2
## Episode: 163 finished with steps:17, rewards:-17.000000 global step 4476
## Last 100 episodes average reward -44.504950
## Epsilon0.059250
## rand steps:1
## Episode: 164 finished with steps:17, rewards:-17.000000 global step 4493
## Last 100 episodes average reward -42.495050
## Epsilon0.059000
## rand steps:2
## Episode: 165 finished with steps:13, rewards:-13.000000 global step 4506
## Last 100 episodes average reward -42.495050
## Epsilon0.058750
## rand steps:0
## Episode: 166 finished with steps:15, rewards:-15.000000 global step 4521
## Last 100 episodes average reward -41.504950
## Epsilon0.058500
## rand steps:1
## Episode: 167 finished with steps:15, rewards:-15.000000 global step 4536
## Last 100 episodes average reward -41.524752
## Epsilon0.058250
## rand steps:2
## Episode: 168 finished with steps:15, rewards:-15.000000 global step 4551
## Last 100 episodes average reward -40.504950
## Epsilon0.058000
## rand steps:1
## Episode: 169 finished with steps:15, rewards:-15.000000 global step 4566
## Last 100 episodes average reward -40.524752
## Epsilon0.057750
## rand steps:1
## Episode: 170 finished with steps:14, rewards:-14.000000 global step 4580
## Last 100 episodes average reward -40.534653
## Epsilon0.057500
## rand steps:1
## Episode: 171 finished with steps:13, rewards:-13.000000 global step 4593
## Last 100 episodes average reward -39.485149
## Epsilon0.057250
## rand steps:1
## Episode: 172 finished with steps:25, rewards:-124.000000 global step 4618
## Last 100 episodes average reward -40.564356
## Epsilon0.057000
## rand steps:1
## Episode: 173 finished with steps:13, rewards:-13.000000 global step 4631
## Last 100 episodes average reward -39.366337
## Epsilon0.056750
## rand steps:0
## Episode: 174 finished with steps:18, rewards:-117.000000 global step 4649
## Last 100 episodes average reward -39.326733
## Epsilon0.056500
## rand steps:2
## Episode: 175 finished with steps:15, rewards:-15.000000 global step 4664
## Last 100 episodes average reward -37.227723
## Epsilon0.056250
## rand steps:1
## Episode: 176 finished with steps:13, rewards:-13.000000 global step 4677
## Last 100 episodes average reward -37.207921
## Epsilon0.056000
## rand steps:0
## Episode: 177 finished with steps:15, rewards:-15.000000 global step 4692
## Last 100 episodes average reward -37.227723
## Epsilon0.055750
## rand steps:2
## Episode: 178 finished with steps:13, rewards:-13.000000 global step 4705
## Last 100 episodes average reward -37.227723
## Epsilon0.055500
## rand steps:0
## Episode: 179 finished with steps:13, rewards:-13.000000 global step 4718
## Last 100 episodes average reward -37.227723
## Epsilon0.055250
## rand steps:1
## Episode: 180 finished with steps:15, rewards:-15.000000 global step 4733
## Last 100 episodes average reward -37.227723
## Epsilon0.055000
## rand steps:1
## Episode: 181 finished with steps:13, rewards:-13.000000 global step 4746
## Last 100 episodes average reward -37.227723
## Epsilon0.054750
## rand steps:0
## Episode: 182 finished with steps:13, rewards:-13.000000 global step 4759
## Last 100 episodes average reward -37.227723
## Epsilon0.054500
## rand steps:0
## Episode: 183 finished with steps:13, rewards:-13.000000 global step 4772
## Last 100 episodes average reward -36.158416
## Epsilon0.054250
## rand steps:2
## Episode: 184 finished with steps:18, rewards:-117.000000 global step 4790
## Last 100 episodes average reward -37.168317
## Epsilon0.054000
## rand steps:2
## Episode: 185 finished with steps:13, rewards:-13.000000 global step 4803
## Last 100 episodes average reward -37.168317
## Epsilon0.053750
## rand steps:1
## Episode: 186 finished with steps:13, rewards:-13.000000 global step 4816
## Last 100 episodes average reward -37.128713
## Epsilon0.053500
## rand steps:0
## Episode: 187 finished with steps:13, rewards:-13.000000 global step 4829
## Last 100 episodes average reward -37.128713
## Epsilon0.053250
## rand steps:0
## Episode: 188 finished with steps:13, rewards:-13.000000 global step 4842
## Last 100 episodes average reward -37.089109
## Epsilon0.053000
## rand steps:1
## Episode: 189 finished with steps:15, rewards:-15.000000 global step 4857
## Last 100 episodes average reward -37.079208
## Epsilon0.052750
## rand steps:1
## Episode: 190 finished with steps:13, rewards:-13.000000 global step 4870
## Last 100 episodes average reward -34.940594
## Epsilon0.052500
## rand steps:0
## Episode: 191 finished with steps:16, rewards:-115.000000 global step 4886
## Last 100 episodes average reward -35.930693
## Epsilon0.052250
## rand steps:1
## Episode: 192 finished with steps:13, rewards:-13.000000 global step 4899
## Last 100 episodes average reward -35.910891
## Epsilon0.052000
## rand steps:0
## Episode: 193 finished with steps:13, rewards:-13.000000 global step 4912
## Last 100 episodes average reward -35.910891
## Epsilon0.051750
## rand steps:1
## Episode: 194 finished with steps:13, rewards:-13.000000 global step 4925
## Last 100 episodes average reward -35.910891
## Epsilon0.051500
## rand steps:0
## Episode: 195 finished with steps:13, rewards:-13.000000 global step 4938
## Last 100 episodes average reward -35.871287
## Epsilon0.051250
## rand steps:0
## Episode: 196 finished with steps:13, rewards:-13.000000 global step 4951
## Last 100 episodes average reward -35.871287
## Epsilon0.051000
## rand steps:0
## Episode: 197 finished with steps:13, rewards:-13.000000 global step 4964
## Last 100 episodes average reward -35.871287
## Epsilon0.050750
## rand steps:0
## Episode: 198 finished with steps:13, rewards:-13.000000 global step 4977
## Last 100 episodes average reward -34.782178
## Epsilon0.050500
## rand steps:0
## Episode: 199 finished with steps:15, rewards:-15.000000 global step 4992
## Last 100 episodes average reward -34.782178
## Epsilon0.050250
## rand steps:1
## Episode: 200 finished with steps:13, rewards:-13.000000 global step 5005
## Last 100 episodes average reward -32.643564
## Epsilon0.050000
## rand steps:0
## Episode: 201 finished with steps:13, rewards:-13.000000 global step 5018
## Last 100 episodes average reward -31.603960
## Epsilon0.049750
## rand steps:0
## Episode: 202 finished with steps:13, rewards:-13.000000 global step 5031
## Last 100 episodes average reward -30.514851
## Epsilon0.049500
## rand steps:0
## Episode: 203 finished with steps:13, rewards:-13.000000 global step 5044
## Last 100 episodes average reward -29.445545
## Epsilon0.049250
## rand steps:0
## Episode: 204 finished with steps:13, rewards:-13.000000 global step 5057
## Last 100 episodes average reward -29.425743
## Epsilon0.049000
## rand steps:0
## Episode: 205 finished with steps:13, rewards:-13.000000 global step 5070
## Last 100 episodes average reward -28.346535
## Epsilon0.048750
## rand steps:0
## Episode: 206 finished with steps:13, rewards:-13.000000 global step 5083
## Last 100 episodes average reward -28.346535
## Epsilon0.048500
## rand steps:0
## Episode: 207 finished with steps:15, rewards:-15.000000 global step 5098
## Last 100 episodes average reward -28.346535
## Epsilon0.048250
## rand steps:1
## Episode: 208 finished with steps:19, rewards:-19.000000 global step 5117
## Last 100 episodes average reward -28.405941
## Epsilon0.048000
## rand steps:3
## Episode: 209 finished with steps:13, rewards:-13.000000 global step 5130
## Last 100 episodes average reward -28.405941
## Epsilon0.047750
## rand steps:0
## Episode: 210 finished with steps:25, rewards:-124.000000 global step 5155
## Last 100 episodes average reward -28.376238
## Epsilon0.047500
## rand steps:1
## Episode: 211 finished with steps:17, rewards:-17.000000 global step 5172
## Last 100 episodes average reward -28.415842
## Epsilon0.047250
## rand steps:3
## Episode: 212 finished with steps:13, rewards:-13.000000 global step 5185
## Last 100 episodes average reward -28.415842
## Epsilon0.047000
## rand steps:1
## Episode: 213 finished with steps:15, rewards:-15.000000 global step 5200
## Last 100 episodes average reward -28.425743
## Epsilon0.046750
## rand steps:1
## Episode: 214 finished with steps:13, rewards:-13.000000 global step 5213
## Last 100 episodes average reward -27.415842
## Epsilon0.046500
## rand steps:0
## Episode: 215 finished with steps:15, rewards:-15.000000 global step 5228
## Last 100 episodes average reward -27.425743
## Epsilon0.046250
## rand steps:1
## Episode: 216 finished with steps:32, rewards:-230.000000 global step 5260
## Last 100 episodes average reward -28.554455
## Epsilon0.046000
## rand steps:5
## Episode: 217 finished with steps:13, rewards:-13.000000 global step 5273
## Last 100 episodes average reward -26.485149
## Epsilon0.045750
## rand steps:0
## Episode: 218 finished with steps:13, rewards:-13.000000 global step 5286
## Last 100 episodes average reward -26.485149
## Epsilon0.045500
## rand steps:0
## Episode: 219 finished with steps:14, rewards:-14.000000 global step 5300
## Last 100 episodes average reward -26.495050
## Epsilon0.045250
## rand steps:2
## Episode: 220 finished with steps:13, rewards:-13.000000 global step 5313
## Last 100 episodes average reward -26.495050
## Epsilon0.045000
## rand steps:0
## Episode: 221 finished with steps:13, rewards:-13.000000 global step 5326
## Last 100 episodes average reward -26.495050
## Epsilon0.044750
## rand steps:0
## Episode: 222 finished with steps:13, rewards:-13.000000 global step 5339
## Last 100 episodes average reward -26.475248
## Epsilon0.044500
## rand steps:0
## Episode: 223 finished with steps:13, rewards:-13.000000 global step 5352
## Last 100 episodes average reward -26.475248
## Epsilon0.044250
## rand steps:0
## Episode: 224 finished with steps:13, rewards:-13.000000 global step 5365
## Last 100 episodes average reward -26.435644
## Epsilon0.044000
## rand steps:1
## Episode: 225 finished with steps:13, rewards:-13.000000 global step 5378
## Last 100 episodes average reward -26.435644
## Epsilon0.043750
## rand steps:0
## Episode: 226 finished with steps:15, rewards:-15.000000 global step 5393
## Last 100 episodes average reward -26.386139
## Epsilon0.043500
## rand steps:1
## Episode: 227 finished with steps:17, rewards:-17.000000 global step 5410
## Last 100 episodes average reward -26.425743
## Epsilon0.043250
## rand steps:2
## Episode: 228 finished with steps:18, rewards:-18.000000 global step 5428
## Last 100 episodes average reward -26.346535
## Epsilon0.043000
## rand steps:3
## Episode: 229 finished with steps:13, rewards:-13.000000 global step 5441
## Last 100 episodes average reward -25.267327
## Epsilon0.042750
## rand steps:0
## Episode: 230 finished with steps:14, rewards:-14.000000 global step 5455
## Last 100 episodes average reward -25.277228
## Epsilon0.042500
## rand steps:1
## Episode: 231 finished with steps:13, rewards:-13.000000 global step 5468
## Last 100 episodes average reward -25.277228
## Epsilon0.042250
## rand steps:0
## Episode: 232 finished with steps:13, rewards:-13.000000 global step 5481
## Last 100 episodes average reward -25.277228
## Epsilon0.042000
## rand steps:0
## Episode: 233 finished with steps:13, rewards:-13.000000 global step 5494
## Last 100 episodes average reward -25.277228
## Epsilon0.041750
## rand steps:0
## Episode: 234 finished with steps:13, rewards:-13.000000 global step 5507
## Last 100 episodes average reward -25.277228
## Epsilon0.041500
## rand steps:0
## Episode: 235 finished with steps:13, rewards:-13.000000 global step 5520
## Last 100 episodes average reward -25.277228
## Epsilon0.041250
## rand steps:0
## Episode: 236 finished with steps:13, rewards:-13.000000 global step 5533
## Last 100 episodes average reward -25.277228
## Epsilon0.041000
## rand steps:0
## Episode: 237 finished with steps:13, rewards:-13.000000 global step 5546
## Last 100 episodes average reward -25.277228
## Epsilon0.040750
## rand steps:0
## Episode: 238 finished with steps:13, rewards:-13.000000 global step 5559
## Last 100 episodes average reward -25.277228
## Epsilon0.040500
## rand steps:0
## Episode: 239 finished with steps:13, rewards:-13.000000 global step 5572
## Last 100 episodes average reward -25.277228
## Epsilon0.040250
## rand steps:0
## Episode: 240 finished with steps:15, rewards:-15.000000 global step 5587
## Last 100 episodes average reward -24.217822
## Epsilon0.040000
## rand steps:1
## Episode: 241 finished with steps:17, rewards:-17.000000 global step 5604
## Last 100 episodes average reward -24.257426
## Epsilon0.039750
## rand steps:1
## Episode: 242 finished with steps:13, rewards:-13.000000 global step 5617
## Last 100 episodes average reward -23.237624
## Epsilon0.039500
## rand steps:0
## Episode: 243 finished with steps:13, rewards:-13.000000 global step 5630
## Last 100 episodes average reward -23.237624
## Epsilon0.039250
## rand steps:0
## Episode: 244 finished with steps:13, rewards:-13.000000 global step 5643
## Last 100 episodes average reward -23.237624
## Epsilon0.039000
## rand steps:0
## Episode: 245 finished with steps:13, rewards:-13.000000 global step 5656
## Last 100 episodes average reward -23.198020
## Epsilon0.038750
## rand steps:0
## Episode: 246 finished with steps:33, rewards:-231.000000 global step 5689
## Last 100 episodes average reward -25.356436
## Epsilon0.038500
## rand steps:4
## Episode: 247 finished with steps:25, rewards:-124.000000 global step 5714
## Last 100 episodes average reward -26.435644
## Epsilon0.038250
## rand steps:2
## Episode: 248 finished with steps:13, rewards:-13.000000 global step 5727
## Last 100 episodes average reward -26.435644
## Epsilon0.038000
## rand steps:0
## Episode: 249 finished with steps:13, rewards:-13.000000 global step 5740
## Last 100 episodes average reward -26.435644
## Epsilon0.037750
## rand steps:0
## Episode: 250 finished with steps:24, rewards:-123.000000 global step 5764
## Last 100 episodes average reward -27.524752
## Epsilon0.037500
## rand steps:1
## Episode: 251 finished with steps:13, rewards:-13.000000 global step 5777
## Last 100 episodes average reward -27.475248
## Epsilon0.037250
## rand steps:0
## Episode: 252 finished with steps:15, rewards:-15.000000 global step 5792
## Last 100 episodes average reward -27.495050
## Epsilon0.037000
## rand steps:1
## Episode: 253 finished with steps:15, rewards:-15.000000 global step 5807
## Last 100 episodes average reward -27.514851
## Epsilon0.036750
## rand steps:1
## Episode: 254 finished with steps:21, rewards:-120.000000 global step 5828
## Last 100 episodes average reward -28.554455
## Epsilon0.036500
## rand steps:2
## Episode: 255 finished with steps:13, rewards:-13.000000 global step 5841
## Last 100 episodes average reward -28.554455
## Epsilon0.036250
## rand steps:0
## Episode: 256 finished with steps:13, rewards:-13.000000 global step 5854
## Last 100 episodes average reward -28.534653
## Epsilon0.036000
## rand steps:0
## Episode: 257 finished with steps:13, rewards:-13.000000 global step 5867
## Last 100 episodes average reward -28.534653
## Epsilon0.035750
## rand steps:0
## Episode: 258 finished with steps:13, rewards:-13.000000 global step 5880
## Last 100 episodes average reward -28.534653
## Epsilon0.035500
## rand steps:2
## Episode: 259 finished with steps:15, rewards:-15.000000 global step 5895
## Last 100 episodes average reward -28.554455
## Epsilon0.035250
## rand steps:1
## Episode: 260 finished with steps:13, rewards:-13.000000 global step 5908
## Last 100 episodes average reward -28.534653
## Epsilon0.035000
## rand steps:0
## Episode: 261 finished with steps:27, rewards:-27.000000 global step 5935
## Last 100 episodes average reward -28.673267
## Epsilon0.034750
## rand steps:2
## Episode: 262 finished with steps:24, rewards:-123.000000 global step 5959
## Last 100 episodes average reward -29.722772
## Epsilon0.034500
## rand steps:1
## Episode: 263 finished with steps:13, rewards:-13.000000 global step 5972
## Last 100 episodes average reward -27.722772
## Epsilon0.034250
## rand steps:0
## Episode: 264 finished with steps:13, rewards:-13.000000 global step 5985
## Last 100 episodes average reward -27.683168
## Epsilon0.034000
## rand steps:1
## Episode: 265 finished with steps:13, rewards:-13.000000 global step 5998
## Last 100 episodes average reward -27.643564
## Epsilon0.033750
## rand steps:0
## Episode: 266 finished with steps:13, rewards:-13.000000 global step 6011
## Last 100 episodes average reward -27.643564
## Epsilon0.033500
## rand steps:0
## Episode: 267 finished with steps:13, rewards:-13.000000 global step 6024
## Last 100 episodes average reward -27.623762
## Epsilon0.033250
## rand steps:0
## Episode: 268 finished with steps:13, rewards:-13.000000 global step 6037
## Last 100 episodes average reward -27.603960
## Epsilon0.033000
## rand steps:0
## Episode: 269 finished with steps:15, rewards:-15.000000 global step 6052
## Last 100 episodes average reward -27.603960
## Epsilon0.032750
## rand steps:2
## Episode: 270 finished with steps:13, rewards:-13.000000 global step 6065
## Last 100 episodes average reward -27.584158
## Epsilon0.032500
## rand steps:0
## Episode: 271 finished with steps:17, rewards:-116.000000 global step 6082
## Last 100 episodes average reward -28.594059
## Epsilon0.032250
## rand steps:2
## Episode: 272 finished with steps:13, rewards:-13.000000 global step 6095
## Last 100 episodes average reward -28.594059
## Epsilon0.032000
## rand steps:0
## Episode: 273 finished with steps:19, rewards:-19.000000 global step 6114
## Last 100 episodes average reward -27.554455
## Epsilon0.031750
## rand steps:3
## Episode: 274 finished with steps:13, rewards:-13.000000 global step 6127
## Last 100 episodes average reward -27.554455
## Epsilon0.031500
## rand steps:0
## Episode: 275 finished with steps:13, rewards:-13.000000 global step 6140
## Last 100 episodes average reward -26.524752
## Epsilon0.031250
## rand steps:0
## Episode: 276 finished with steps:15, rewards:-15.000000 global step 6155
## Last 100 episodes average reward -26.524752
## Epsilon0.031000
## rand steps:2
## Episode: 277 finished with steps:13, rewards:-13.000000 global step 6168
## Last 100 episodes average reward -26.524752
## Epsilon0.030750
## rand steps:0
## Episode: 278 finished with steps:20, rewards:-119.000000 global step 6188
## Last 100 episodes average reward -27.554455
## Epsilon0.030500
## rand steps:1
## Episode: 279 finished with steps:13, rewards:-13.000000 global step 6201
## Last 100 episodes average reward -27.554455
## Epsilon0.030250
## rand steps:0
## Episode: 280 finished with steps:13, rewards:-13.000000 global step 6214
## Last 100 episodes average reward -27.554455
## Epsilon0.030000
## rand steps:0
## Episode: 281 finished with steps:13, rewards:-13.000000 global step 6227
## Last 100 episodes average reward -27.534653
## Epsilon0.029750
## rand steps:0
## Episode: 282 finished with steps:13, rewards:-13.000000 global step 6240
## Last 100 episodes average reward -27.534653
## Epsilon0.029500
## rand steps:0
## Episode: 283 finished with steps:13, rewards:-13.000000 global step 6253
## Last 100 episodes average reward -27.534653
## Epsilon0.029250
## rand steps:0
## Episode: 284 finished with steps:16, rewards:-115.000000 global step 6269
## Last 100 episodes average reward -28.544554
## Epsilon0.029000
## rand steps:1
## Episode: 285 finished with steps:13, rewards:-13.000000 global step 6282
## Last 100 episodes average reward -27.514851
## Epsilon0.028750
## rand steps:0
## Episode: 286 finished with steps:13, rewards:-13.000000 global step 6295
## Last 100 episodes average reward -27.514851
## Epsilon0.028500
## rand steps:0
## Episode: 287 finished with steps:13, rewards:-13.000000 global step 6308
## Last 100 episodes average reward -27.514851
## Epsilon0.028250
## rand steps:0
## Episode: 288 finished with steps:13, rewards:-13.000000 global step 6321
## Last 100 episodes average reward -27.514851
## Epsilon0.028000
## rand steps:0
## Episode: 289 finished with steps:13, rewards:-13.000000 global step 6334
## Last 100 episodes average reward -27.514851
## Epsilon0.027750
## rand steps:0
## Episode: 290 finished with steps:13, rewards:-13.000000 global step 6347
## Last 100 episodes average reward -27.495050
## Epsilon0.027500
## rand steps:0
## Episode: 291 finished with steps:15, rewards:-15.000000 global step 6362
## Last 100 episodes average reward -27.514851
## Epsilon0.027250
## rand steps:1
## Episode: 292 finished with steps:13, rewards:-13.000000 global step 6375
## Last 100 episodes average reward -26.504950
## Epsilon0.027000
## rand steps:0
## Episode: 293 finished with steps:13, rewards:-13.000000 global step 6388
## Last 100 episodes average reward -26.504950
## Epsilon0.026750
## rand steps:0
## Episode: 294 finished with steps:13, rewards:-13.000000 global step 6401
## Last 100 episodes average reward -26.504950
## Epsilon0.026500
## rand steps:0
## Episode: 295 finished with steps:13, rewards:-13.000000 global step 6414
## Last 100 episodes average reward -26.504950
## Epsilon0.026250
## rand steps:0
## Episode: 296 finished with steps:15, rewards:-15.000000 global step 6429
## Last 100 episodes average reward -26.524752
## Epsilon0.026000
## rand steps:1
## Episode: 297 finished with steps:15, rewards:-15.000000 global step 6444
## Last 100 episodes average reward -26.544554
## Epsilon0.025750
## rand steps:1
## Episode: 298 finished with steps:13, rewards:-13.000000 global step 6457
## Last 100 episodes average reward -26.544554
## Epsilon0.025500
## rand steps:1
## Episode: 299 finished with steps:13, rewards:-13.000000 global step 6470
## Last 100 episodes average reward -26.544554
## Epsilon0.025250
## rand steps:0
## Episode: 300 finished with steps:13, rewards:-13.000000 global step 6483
## Last 100 episodes average reward -26.524752
## Epsilon0.025000
## rand steps:0
## Episode: 301 finished with steps:15, rewards:-15.000000 global step 6498
## Last 100 episodes average reward -26.544554
## Epsilon0.024750
## rand steps:1
## Episode: 302 finished with steps:13, rewards:-13.000000 global step 6511
## Last 100 episodes average reward -26.544554
## Epsilon0.024500
## rand steps:0
## Episode: 303 finished with steps:13, rewards:-13.000000 global step 6524
## Last 100 episodes average reward -26.544554
## Epsilon0.024250
## rand steps:0
## Episode: 304 finished with steps:13, rewards:-13.000000 global step 6537
## Last 100 episodes average reward -26.544554
## Epsilon0.024000
## rand steps:1
## Episode: 305 finished with steps:13, rewards:-13.000000 global step 6550
## Last 100 episodes average reward -26.544554
## Epsilon0.023750
## rand steps:0
## Episode: 306 finished with steps:13, rewards:-13.000000 global step 6563
## Last 100 episodes average reward -26.544554
## Epsilon0.023500
## rand steps:0
## Episode: 307 finished with steps:13, rewards:-13.000000 global step 6576
## Last 100 episodes average reward -26.544554
## Epsilon0.023250
## rand steps:1
## Episode: 308 finished with steps:13, rewards:-13.000000 global step 6589
## Last 100 episodes average reward -26.524752
## Epsilon0.023000
## rand steps:0
## Episode: 309 finished with steps:24, rewards:-123.000000 global step 6613
## Last 100 episodes average reward -27.554455
## Epsilon0.022750
## rand steps:2
## Episode: 310 finished with steps:13, rewards:-13.000000 global step 6626
## Last 100 episodes average reward -27.554455
## Epsilon0.022500
## rand steps:0
## Episode: 311 finished with steps:13, rewards:-13.000000 global step 6639
## Last 100 episodes average reward -26.455446
## Epsilon0.022250
## rand steps:0
## Episode: 312 finished with steps:15, rewards:-15.000000 global step 6654
## Last 100 episodes average reward -26.435644
## Epsilon0.022000
## rand steps:1
## Episode: 313 finished with steps:13, rewards:-13.000000 global step 6667
## Last 100 episodes average reward -26.435644
## Epsilon0.021750
## rand steps:0
## Episode: 314 finished with steps:13, rewards:-13.000000 global step 6680
## Last 100 episodes average reward -26.415842
## Epsilon0.021500
## rand steps:0
## Episode: 315 finished with steps:13, rewards:-13.000000 global step 6693
## Last 100 episodes average reward -26.415842
## Epsilon0.021250
## rand steps:0
## Episode: 316 finished with steps:13, rewards:-13.000000 global step 6706
## Last 100 episodes average reward -26.396040
## Epsilon0.021000
## rand steps:0
## Episode: 317 finished with steps:13, rewards:-13.000000 global step 6719
## Last 100 episodes average reward -24.247525
## Epsilon0.020750
## rand steps:0
## Episode: 318 finished with steps:13, rewards:-13.000000 global step 6732
## Last 100 episodes average reward -24.247525
## Epsilon0.020500
## rand steps:0
## Episode: 319 finished with steps:13, rewards:-13.000000 global step 6745
## Last 100 episodes average reward -24.247525
## Epsilon0.020250
## rand steps:0
## Episode: 320 finished with steps:13, rewards:-13.000000 global step 6758
## Last 100 episodes average reward -24.237624
## Epsilon0.020000
## rand steps:0
## Episode: 321 finished with steps:26, rewards:-224.000000 global step 6784
## Last 100 episodes average reward -26.326733
## Epsilon0.019750
## rand steps:2
## Episode: 322 finished with steps:13, rewards:-13.000000 global step 6797
## Last 100 episodes average reward -26.326733
## Epsilon0.019500
## rand steps:0
## Episode: 323 finished with steps:13, rewards:-13.000000 global step 6810
## Last 100 episodes average reward -26.326733
## Epsilon0.019250
## rand steps:0
## Episode: 324 finished with steps:25, rewards:-124.000000 global step 6835
## Last 100 episodes average reward -27.425743
## Epsilon0.019000
## rand steps:1
## Episode: 325 finished with steps:20, rewards:-119.000000 global step 6855
## Last 100 episodes average reward -28.475248
## Epsilon0.018750
## rand steps:1
## Episode: 326 finished with steps:15, rewards:-15.000000 global step 6870
## Last 100 episodes average reward -28.495050
## Epsilon0.018500
## rand steps:1
## Episode: 327 finished with steps:13, rewards:-13.000000 global step 6883
## Last 100 episodes average reward -28.475248
## Epsilon0.018250
## rand steps:0
## Episode: 328 finished with steps:13, rewards:-13.000000 global step 6896
## Last 100 episodes average reward -28.435644
## Epsilon0.018000
## rand steps:0
## Episode: 329 finished with steps:13, rewards:-13.000000 global step 6909
## Last 100 episodes average reward -28.386139
## Epsilon0.017750
## rand steps:0
## Episode: 330 finished with steps:13, rewards:-13.000000 global step 6922
## Last 100 episodes average reward -28.386139
## Epsilon0.017500
## rand steps:0
## Episode: 331 finished with steps:13, rewards:-13.000000 global step 6935
## Last 100 episodes average reward -28.376238
## Epsilon0.017250
## rand steps:1
## Episode: 332 finished with steps:13, rewards:-13.000000 global step 6948
## Last 100 episodes average reward -28.376238
## Epsilon0.017000
## rand steps:0
## Episode: 333 finished with steps:13, rewards:-13.000000 global step 6961
## Last 100 episodes average reward -28.376238
## Epsilon0.016750
## rand steps:0
## Episode: 334 finished with steps:21, rewards:-21.000000 global step 6982
## Last 100 episodes average reward -28.455446
## Epsilon0.016500
## rand steps:2
## Episode: 335 finished with steps:19, rewards:-118.000000 global step 7001
## Last 100 episodes average reward -29.495050
## Epsilon0.016250
## rand steps:1
## Episode: 336 finished with steps:13, rewards:-13.000000 global step 7014
## Last 100 episodes average reward -29.495050
## Epsilon0.016000
## rand steps:0
## Episode: 337 finished with steps:13, rewards:-13.000000 global step 7027
## Last 100 episodes average reward -29.495050
## Epsilon0.015750
## rand steps:0
## Episode: 338 finished with steps:15, rewards:-15.000000 global step 7042
## Last 100 episodes average reward -29.514851
## Epsilon0.015500
## rand steps:1
## Episode: 339 finished with steps:13, rewards:-13.000000 global step 7055
## Last 100 episodes average reward -29.514851
## Epsilon0.015250
## rand steps:0
## Episode: 340 finished with steps:13, rewards:-13.000000 global step 7068
## Last 100 episodes average reward -29.514851
## Epsilon0.015000
## rand steps:0
## Episode: 341 finished with steps:13, rewards:-13.000000 global step 7081
## Last 100 episodes average reward -29.495050
## Epsilon0.014750
## rand steps:0
## Episode: 342 finished with steps:13, rewards:-13.000000 global step 7094
## Last 100 episodes average reward -29.455446
## Epsilon0.014500
## rand steps:0
## Episode: 343 finished with steps:13, rewards:-13.000000 global step 7107
## Last 100 episodes average reward -29.455446
## Epsilon0.014250
## rand steps:0
## Episode: 344 finished with steps:13, rewards:-13.000000 global step 7120
## Last 100 episodes average reward -29.455446
## Epsilon0.014000
## rand steps:1
## Episode: 345 finished with steps:13, rewards:-13.000000 global step 7133
## Last 100 episodes average reward -29.455446
## Epsilon0.013750
## rand steps:0
## Episode: 346 finished with steps:13, rewards:-13.000000 global step 7146
## Last 100 episodes average reward -29.455446
## Epsilon0.013500
## rand steps:0
## Episode: 347 finished with steps:13, rewards:-13.000000 global step 7159
## Last 100 episodes average reward -27.297030
## Epsilon0.013250
## rand steps:0
## Episode: 348 finished with steps:13, rewards:-13.000000 global step 7172
## Last 100 episodes average reward -26.198020
## Epsilon0.013000
## rand steps:0
## Episode: 349 finished with steps:13, rewards:-13.000000 global step 7185
## Last 100 episodes average reward -26.198020
## Epsilon0.012750
## rand steps:0
## Episode: 350 finished with steps:13, rewards:-13.000000 global step 7198
## Last 100 episodes average reward -26.198020
## Epsilon0.012500
## rand steps:0
## Episode: 351 finished with steps:15, rewards:-15.000000 global step 7213
## Last 100 episodes average reward -25.128713
## Epsilon0.012250
## rand steps:2
## Episode: 352 finished with steps:15, rewards:-15.000000 global step 7228
## Last 100 episodes average reward -25.148515
## Epsilon0.012000
## rand steps:1
## Episode: 353 finished with steps:13, rewards:-13.000000 global step 7241
## Last 100 episodes average reward -25.128713
## Epsilon0.011750
## rand steps:0
## Episode: 354 finished with steps:13, rewards:-13.000000 global step 7254
## Last 100 episodes average reward -25.108911
## Epsilon0.011500
## rand steps:0
## Episode: 355 finished with steps:13, rewards:-13.000000 global step 7267
## Last 100 episodes average reward -24.049505
## Epsilon0.011250
## rand steps:0
## Episode: 356 finished with steps:13, rewards:-13.000000 global step 7280
## Last 100 episodes average reward -24.049505
## Epsilon0.011000
## rand steps:0
## Episode: 357 finished with steps:13, rewards:-13.000000 global step 7293
## Last 100 episodes average reward -24.049505
## Epsilon0.010750
## rand steps:0
## Episode: 358 finished with steps:13, rewards:-13.000000 global step 7306
## Last 100 episodes average reward -24.049505
## Epsilon0.010500
## rand steps:0
## Episode: 359 finished with steps:13, rewards:-13.000000 global step 7319
## Last 100 episodes average reward -24.049505
## Epsilon0.010250
## rand steps:0
## Episode: 360 finished with steps:13, rewards:-13.000000 global step 7332
## Last 100 episodes average reward -24.029703
## Epsilon0.010000
## rand steps:0
## Episode: 361 finished with steps:13, rewards:-13.000000 global step 7345
## Last 100 episodes average reward -24.029703
## Epsilon0.009750
## rand steps:0
## Episode: 362 finished with steps:13, rewards:-13.000000 global step 7358
## Last 100 episodes average reward -23.891089
## Epsilon0.009500
## rand steps:0
## Episode: 363 finished with steps:13, rewards:-13.000000 global step 7371
## Last 100 episodes average reward -22.801980
## Epsilon0.009250
## rand steps:0
## Episode: 364 finished with steps:13, rewards:-13.000000 global step 7384
## Last 100 episodes average reward -22.801980
## Epsilon0.009000
## rand steps:0
## Episode: 365 finished with steps:13, rewards:-13.000000 global step 7397
## Last 100 episodes average reward -22.801980
## Epsilon0.008750
## rand steps:0
## Episode: 366 finished with steps:15, rewards:-15.000000 global step 7412
## Last 100 episodes average reward -22.821782
## Epsilon0.008500
## rand steps:1
## Episode: 367 finished with steps:13, rewards:-13.000000 global step 7425
## Last 100 episodes average reward -22.821782
## Epsilon0.008250
## rand steps:0
## Episode: 368 finished with steps:13, rewards:-13.000000 global step 7438
## Last 100 episodes average reward -22.821782
## Epsilon0.008000
## rand steps:0
## Episode: 369 finished with steps:13, rewards:-13.000000 global step 7451
## Last 100 episodes average reward -22.821782
## Epsilon0.007750
## rand steps:0
## Episode: 370 finished with steps:13, rewards:-13.000000 global step 7464
## Last 100 episodes average reward -22.801980
## Epsilon0.007500
## rand steps:0
## Episode: 371 finished with steps:13, rewards:-13.000000 global step 7477
## Last 100 episodes average reward -22.801980
## Epsilon0.007250
## rand steps:0
## Episode: 372 finished with steps:13, rewards:-13.000000 global step 7490
## Last 100 episodes average reward -21.782178
## Epsilon0.007000
## rand steps:0
## Episode: 373 finished with steps:13, rewards:-13.000000 global step 7503
## Last 100 episodes average reward -21.782178
## Epsilon0.006750
## rand steps:0
## Episode: 374 finished with steps:15, rewards:-15.000000 global step 7518
## Last 100 episodes average reward -21.742574
## Epsilon0.006500
## rand steps:1
## Episode: 375 finished with steps:13, rewards:-13.000000 global step 7531
## Last 100 episodes average reward -21.742574
## Epsilon0.006250
## rand steps:0
## Episode: 376 finished with steps:13, rewards:-13.000000 global step 7544
## Last 100 episodes average reward -21.742574
## Epsilon0.006000
## rand steps:0
## Episode: 377 finished with steps:13, rewards:-13.000000 global step 7557
## Last 100 episodes average reward -21.722772
## Epsilon0.005750
## rand steps:0
## Episode: 378 finished with steps:13, rewards:-13.000000 global step 7570
## Last 100 episodes average reward -21.722772
## Epsilon0.005500
## rand steps:0
## Episode: 379 finished with steps:13, rewards:-13.000000 global step 7583
## Last 100 episodes average reward -20.673267
## Epsilon0.005250
## rand steps:0
## Episode: 380 finished with steps:13, rewards:-13.000000 global step 7596
## Last 100 episodes average reward -20.673267
## Epsilon0.005000
## rand steps:0
## Episode: 381 finished with steps:13, rewards:-13.000000 global step 7609
## Last 100 episodes average reward -20.673267
## Epsilon0.004750
## rand steps:0
## Episode: 382 finished with steps:13, rewards:-13.000000 global step 7622
## Last 100 episodes average reward -20.673267
## Epsilon0.004500
## rand steps:0
## Episode: 383 finished with steps:13, rewards:-13.000000 global step 7635
## Last 100 episodes average reward -20.673267
## Epsilon0.004250
## rand steps:0
## Episode: 384 finished with steps:13, rewards:-13.000000 global step 7648
## Last 100 episodes average reward -20.673267
## Epsilon0.004000
## rand steps:0
## Episode: 385 finished with steps:13, rewards:-13.000000 global step 7661
## Last 100 episodes average reward -19.663366
## Epsilon0.003750
## rand steps:0
## Episode: 386 finished with steps:13, rewards:-13.000000 global step 7674
## Last 100 episodes average reward -19.663366
## Epsilon0.003500
## rand steps:0
## Episode: 387 finished with steps:13, rewards:-13.000000 global step 7687
## Last 100 episodes average reward -19.663366
## Epsilon0.003250
## rand steps:0
## Episode: 388 finished with steps:13, rewards:-13.000000 global step 7700
## Last 100 episodes average reward -19.663366
## Epsilon0.003000
## rand steps:0
## Episode: 389 finished with steps:13, rewards:-13.000000 global step 7713
## Last 100 episodes average reward -19.663366
## Epsilon0.002750
## rand steps:0
## Episode: 390 finished with steps:13, rewards:-13.000000 global step 7726
## Last 100 episodes average reward -19.663366
## Epsilon0.002500
## rand steps:0
## Episode: 391 finished with steps:13, rewards:-13.000000 global step 7739
## Last 100 episodes average reward -19.663366
## Epsilon0.002250
## rand steps:0
## Episode: 392 finished with steps:13, rewards:-13.000000 global step 7752
## Last 100 episodes average reward -19.643564
## Epsilon0.002000
## rand steps:0
## Episode: 393 finished with steps:13, rewards:-13.000000 global step 7765
## Last 100 episodes average reward -19.643564
## Epsilon0.001750
## rand steps:0
## Episode: 394 finished with steps:13, rewards:-13.000000 global step 7778
## Last 100 episodes average reward -19.643564
## Epsilon0.001500
## rand steps:0
## Episode: 395 finished with steps:13, rewards:-13.000000 global step 7791
## Last 100 episodes average reward -19.643564
## Epsilon0.001250
## rand steps:0
## Episode: 396 finished with steps:13, rewards:-13.000000 global step 7804
## Last 100 episodes average reward -19.643564
## Epsilon0.001000
## rand steps:0
## Episode: 397 finished with steps:13, rewards:-13.000000 global step 7817
## Last 100 episodes average reward -19.623762
## Epsilon0.000750
## rand steps:0
## Episode: 398 finished with steps:13, rewards:-13.000000 global step 7830
## Last 100 episodes average reward -19.603960
## Epsilon0.000500
## rand steps:0
## Episode: 399 finished with steps:13, rewards:-13.000000 global step 7843
## Last 100 episodes average reward -19.603960
## Epsilon0.000250
## rand steps:0
## Episode: 400 finished with steps:13, rewards:-13.000000 global step 7856
## Last 100 episodes average reward -19.603960
## Epsilon0.000000
## rand steps:0
## Episode: 401 finished with steps:13, rewards:-13.000000 global step 7869
## Last 100 episodes average reward -19.603960
## Epsilon0.000000
## rand steps:0
## Episode: 402 finished with steps:13, rewards:-13.000000 global step 7882
## Last 100 episodes average reward -19.584158
## Epsilon0.000000
## rand steps:0
## Episode: 403 finished with steps:13, rewards:-13.000000 global step 7895
## Last 100 episodes average reward -19.584158
## Epsilon0.000000
## rand steps:0
## Episode: 404 finished with steps:13, rewards:-13.000000 global step 7908
## Last 100 episodes average reward -19.584158
## Epsilon0.000000
## rand steps:0
## Episode: 405 finished with steps:13, rewards:-13.000000 global step 7921
## Last 100 episodes average reward -19.584158
## Epsilon0.000000
## rand steps:0
## Episode: 406 finished with steps:13, rewards:-13.000000 global step 7934
## Last 100 episodes average reward -19.584158
## Epsilon0.000000
## rand steps:0
## Episode: 407 finished with steps:13, rewards:-13.000000 global step 7947
## Last 100 episodes average reward -19.584158
## Epsilon0.000000
## rand steps:0
## Episode: 408 finished with steps:13, rewards:-13.000000 global step 7960
## Last 100 episodes average reward -19.584158
## Epsilon0.000000
## rand steps:0
## Episode: 409 finished with steps:13, rewards:-13.000000 global step 7973
## Last 100 episodes average reward -19.584158
## Epsilon0.000000
## rand steps:0
## Episode: 410 finished with steps:13, rewards:-13.000000 global step 7986
## Last 100 episodes average reward -18.495050
## Epsilon0.000000
## rand steps:0
## Episode: 411 finished with steps:13, rewards:-13.000000 global step 7999
## Last 100 episodes average reward -18.495050
## Epsilon0.000000
## rand steps:0
## Episode: 412 finished with steps:13, rewards:-13.000000 global step 8012
## Last 100 episodes average reward -18.495050
## Epsilon0.000000
## rand steps:0
## Episode: 413 finished with steps:13, rewards:-13.000000 global step 8025
## Last 100 episodes average reward -18.475248
## Epsilon0.000000
## rand steps:0
## Episode: 414 finished with steps:13, rewards:-13.000000 global step 8038
## Last 100 episodes average reward -18.475248
## Epsilon0.000000
## rand steps:0
## Episode: 415 finished with steps:13, rewards:-13.000000 global step 8051
## Last 100 episodes average reward -18.475248
## Epsilon0.000000
## rand steps:0
## Episode: 416 finished with steps:13, rewards:-13.000000 global step 8064
## Last 100 episodes average reward -18.475248
## Epsilon0.000000
## rand steps:0
## Episode: 417 finished with steps:13, rewards:-13.000000 global step 8077
## Last 100 episodes average reward -18.475248
## Epsilon0.000000
## rand steps:0
## Episode: 418 finished with steps:13, rewards:-13.000000 global step 8090
## Last 100 episodes average reward -18.475248
## Epsilon0.000000
## rand steps:0
## Episode: 419 finished with steps:13, rewards:-13.000000 global step 8103
## Last 100 episodes average reward -18.475248
## Epsilon0.000000
## rand steps:0
## Episode: 420 finished with steps:13, rewards:-13.000000 global step 8116
## Last 100 episodes average reward -18.475248
## Epsilon0.000000
## rand steps:0
## Episode: 421 finished with steps:13, rewards:-13.000000 global step 8129
## Last 100 episodes average reward -18.475248
## Epsilon0.000000
## rand steps:0
## Episode: 422 finished with steps:13, rewards:-13.000000 global step 8142
## Last 100 episodes average reward -16.386139
## Epsilon0.000000
## rand steps:0
## Episode: 423 finished with steps:13, rewards:-13.000000 global step 8155
## Last 100 episodes average reward -16.386139
## Epsilon0.000000
## rand steps:0
## Episode: 424 finished with steps:13, rewards:-13.000000 global step 8168
## Last 100 episodes average reward -16.386139
## Epsilon0.000000
## rand steps:0
## Episode: 425 finished with steps:13, rewards:-13.000000 global step 8181
## Last 100 episodes average reward -15.287129
## Epsilon0.000000
## rand steps:0
## Episode: 426 finished with steps:13, rewards:-13.000000 global step 8194
## Last 100 episodes average reward -14.237624
## Epsilon0.000000
## rand steps:0
## Episode: 427 finished with steps:13, rewards:-13.000000 global step 8207
## Last 100 episodes average reward -14.217822
## Epsilon0.000000
## rand steps:0
## Episode: 428 finished with steps:13, rewards:-13.000000 global step 8220
## Last 100 episodes average reward -14.217822
## Epsilon0.000000
## rand steps:0
## Episode: 429 finished with steps:13, rewards:-13.000000 global step 8233
## Last 100 episodes average reward -14.217822
## Epsilon0.000000
## rand steps:0
## Episode: 430 finished with steps:13, rewards:-13.000000 global step 8246
## Last 100 episodes average reward -14.217822
## Epsilon0.000000
## rand steps:0
## Episode: 431 finished with steps:13, rewards:-13.000000 global step 8259
## Last 100 episodes average reward -14.217822
## Epsilon0.000000
## rand steps:0
## Episode: 432 finished with steps:13, rewards:-13.000000 global step 8272
## Last 100 episodes average reward -14.217822
## Epsilon0.000000
## rand steps:0
## Episode: 433 finished with steps:13, rewards:-13.000000 global step 8285
## Last 100 episodes average reward -14.217822
## Epsilon0.000000
## rand steps:0
## Episode: 434 finished with steps:13, rewards:-13.000000 global step 8298
## Last 100 episodes average reward -14.217822
## Epsilon0.000000
## rand steps:0
## Episode: 435 finished with steps:13, rewards:-13.000000 global step 8311
## Last 100 episodes average reward -14.138614
## Epsilon0.000000
## rand steps:0
## Episode: 436 finished with steps:13, rewards:-13.000000 global step 8324
## Last 100 episodes average reward -13.099010
## Epsilon0.000000
## rand steps:0
## Episode: 437 finished with steps:13, rewards:-13.000000 global step 8337
## Last 100 episodes average reward -13.099010
## Epsilon0.000000
## rand steps:0
## Episode: 438 finished with steps:13, rewards:-13.000000 global step 8350
## Last 100 episodes average reward -13.099010
## Epsilon0.000000
## rand steps:0
## Episode: 439 finished with steps:13, rewards:-13.000000 global step 8363
## Last 100 episodes average reward -13.079208
## Epsilon0.000000
## rand steps:0
## Episode: 440 finished with steps:13, rewards:-13.000000 global step 8376
## Last 100 episodes average reward -13.079208
## Epsilon0.000000
## rand steps:0
## Episode: 441 finished with steps:13, rewards:-13.000000 global step 8389
## Last 100 episodes average reward -13.079208
## Epsilon0.000000
## rand steps:0
## Episode: 442 finished with steps:13, rewards:-13.000000 global step 8402
## Last 100 episodes average reward -13.079208
## Epsilon0.000000
## rand steps:0
## Episode: 443 finished with steps:13, rewards:-13.000000 global step 8415
## Last 100 episodes average reward -13.079208
## Epsilon0.000000
## rand steps:0
## Episode: 444 finished with steps:13, rewards:-13.000000 global step 8428
## Last 100 episodes average reward -13.079208
## Epsilon0.000000
## rand steps:0
## Episode: 445 finished with steps:13, rewards:-13.000000 global step 8441
## Last 100 episodes average reward -13.079208
## Epsilon0.000000
## rand steps:0
## Episode: 446 finished with steps:13, rewards:-13.000000 global step 8454
## Last 100 episodes average reward -13.079208
## Epsilon0.000000
## rand steps:0
## Episode: 447 finished with steps:13, rewards:-13.000000 global step 8467
## Last 100 episodes average reward -13.079208
## Epsilon0.000000
## rand steps:0
## Episode: 448 finished with steps:13, rewards:-13.000000 global step 8480
## Last 100 episodes average reward -13.079208
## Epsilon0.000000
## rand steps:0
## Episode: 449 finished with steps:13, rewards:-13.000000 global step 8493
## Last 100 episodes average reward -13.079208
## Epsilon0.000000
## rand steps:0
## Episode: 450 finished with steps:13, rewards:-13.000000 global step 8506
## Last 100 episodes average reward -13.079208
## Epsilon0.000000
## rand steps:0
## Episode: 451 finished with steps:13, rewards:-13.000000 global step 8519
## Last 100 episodes average reward -13.079208
## Epsilon0.000000
## rand steps:0
## Episode: 452 finished with steps:13, rewards:-13.000000 global step 8532
## Last 100 episodes average reward -13.059406
## Epsilon0.000000
## rand steps:0
## Episode: 453 finished with steps:13, rewards:-13.000000 global step 8545
## Last 100 episodes average reward -13.039604
## Epsilon0.000000
## rand steps:0
## Episode: 454 finished with steps:13, rewards:-13.000000 global step 8558
## Last 100 episodes average reward -13.039604
## Epsilon0.000000
## rand steps:0
## Episode: 455 finished with steps:13, rewards:-13.000000 global step 8571
## Last 100 episodes average reward -13.039604
## Epsilon0.000000
## rand steps:0
## Episode: 456 finished with steps:13, rewards:-13.000000 global step 8584
## Last 100 episodes average reward -13.039604
## Epsilon0.000000
## rand steps:0
## Episode: 457 finished with steps:13, rewards:-13.000000 global step 8597
## Last 100 episodes average reward -13.039604
## Epsilon0.000000
## rand steps:0
## Episode: 458 finished with steps:13, rewards:-13.000000 global step 8610
## Last 100 episodes average reward -13.039604
## Epsilon0.000000
## rand steps:0
## Episode: 459 finished with steps:13, rewards:-13.000000 global step 8623
## Last 100 episodes average reward -13.039604
## Epsilon0.000000
## rand steps:0
## Episode: 460 finished with steps:13, rewards:-13.000000 global step 8636
## Last 100 episodes average reward -13.039604
## Epsilon0.000000
## rand steps:0
## Episode: 461 finished with steps:13, rewards:-13.000000 global step 8649
## Last 100 episodes average reward -13.039604
## Epsilon0.000000
## rand steps:0
## Episode: 462 finished with steps:13, rewards:-13.000000 global step 8662
## Last 100 episodes average reward -13.039604
## Epsilon0.000000
## rand steps:0
## Episode: 463 finished with steps:13, rewards:-13.000000 global step 8675
## Last 100 episodes average reward -13.039604
## Epsilon0.000000
## rand steps:0
## Episode: 464 finished with steps:13, rewards:-13.000000 global step 8688
## Last 100 episodes average reward -13.039604
## Epsilon0.000000
## rand steps:0
## Episode: 465 finished with steps:13, rewards:-13.000000 global step 8701
## Last 100 episodes average reward -13.039604
## Epsilon0.000000
## rand steps:0
## Episode: 466 finished with steps:13, rewards:-13.000000 global step 8714
## Last 100 episodes average reward -13.039604
## Epsilon0.000000
## rand steps:0
## Episode: 467 finished with steps:13, rewards:-13.000000 global step 8727
## Last 100 episodes average reward -13.019802
## Epsilon0.000000
## rand steps:0
## Episode: 468 finished with steps:13, rewards:-13.000000 global step 8740
## Last 100 episodes average reward -13.019802
## Epsilon0.000000
## rand steps:0
## Episode: 469 finished with steps:13, rewards:-13.000000 global step 8753
## Last 100 episodes average reward -13.019802
## Epsilon0.000000
## rand steps:0
## Episode: 470 finished with steps:13, rewards:-13.000000 global step 8766
## Last 100 episodes average reward -13.019802
## Epsilon0.000000
## rand steps:0
## Episode: 471 finished with steps:13, rewards:-13.000000 global step 8779
## Last 100 episodes average reward -13.019802
## Epsilon0.000000
## rand steps:0
## Episode: 472 finished with steps:13, rewards:-13.000000 global step 8792
## Last 100 episodes average reward -13.019802
## Epsilon0.000000
## rand steps:0
## Episode: 473 finished with steps:13, rewards:-13.000000 global step 8805
## Last 100 episodes average reward -13.019802
## Epsilon0.000000
## rand steps:0
## Episode: 474 finished with steps:13, rewards:-13.000000 global step 8818
## Last 100 episodes average reward -13.019802
## Epsilon0.000000
## rand steps:0
## Episode: 475 finished with steps:13, rewards:-13.000000 global step 8831
## Last 100 episodes average reward -13.000000
## Epsilon0.000000
## rand steps:0
## Episode: 476 finished with steps:13, rewards:-13.000000 global step 8844
## Last 100 episodes average reward -13.000000
## Epsilon0.000000
## rand steps:0
## Episode: 477 finished with steps:13, rewards:-13.000000 global step 8857
## Last 100 episodes average reward -13.000000
## Epsilon0.000000
## rand steps:0
## Episode: 478 finished with steps:13, rewards:-13.000000 global step 8870
## Last 100 episodes average reward -13.000000
## Epsilon0.000000
## rand steps:0
## Episode: 479 finished with steps:13, rewards:-13.000000 global step 8883
## Last 100 episodes average reward -13.000000
## Epsilon0.000000
## rand steps:0
## Episode: 480 finished with steps:13, rewards:-13.000000 global step 8896
## Last 100 episodes average reward -13.000000
## Epsilon0.000000
## rand steps:0
## Episode: 481 finished with steps:13, rewards:-13.000000 global step 8909
## Last 100 episodes average reward -13.000000
## Epsilon0.000000
## rand steps:0
## Episode: 482 finished with steps:13, rewards:-13.000000 global step 8922
## Last 100 episodes average reward -13.000000
## Epsilon0.000000
## rand steps:0
## Episode: 483 finished with steps:13, rewards:-13.000000 global step 8935
## Last 100 episodes average reward -13.000000
## Epsilon0.000000
## rand steps:0
## Episode: 484 finished with steps:13, rewards:-13.000000 global step 8948
## Last 100 episodes average reward -13.000000
## Epsilon0.000000
## rand steps:0
## Episode: 485 finished with steps:13, rewards:-13.000000 global step 8961
## Last 100 episodes average reward -13.000000
## Epsilon0.000000
## rand steps:0
## Episode: 486 finished with steps:13, rewards:-13.000000 global step 8974
## Last 100 episodes average reward -13.000000
## Epsilon0.000000
## rand steps:0
## Episode: 487 finished with steps:13, rewards:-13.000000 global step 8987
## Last 100 episodes average reward -13.000000
## Epsilon0.000000
## rand steps:0
## Episode: 488 finished with steps:13, rewards:-13.000000 global step 9000
## Last 100 episodes average reward -13.000000
## Epsilon0.000000
## rand steps:0
## Episode: 489 finished with steps:13, rewards:-13.000000 global step 9013
## Last 100 episodes average reward -13.000000
## Epsilon0.000000
## rand steps:0
## Episode: 490 finished with steps:13, rewards:-13.000000 global step 9026
## Last 100 episodes average reward -13.000000
## Epsilon0.000000
## rand steps:0
## Episode: 491 finished with steps:13, rewards:-13.000000 global step 9039
## Last 100 episodes average reward -13.000000
## Epsilon0.000000
## rand steps:0
## Episode: 492 finished with steps:13, rewards:-13.000000 global step 9052
## Last 100 episodes average reward -13.000000
## Epsilon0.000000
## rand steps:0
## Episode: 493 finished with steps:13, rewards:-13.000000 global step 9065
## Last 100 episodes average reward -13.000000
## Epsilon0.000000
## rand steps:0
## Episode: 494 finished with steps:13, rewards:-13.000000 global step 9078
## Last 100 episodes average reward -13.000000
## Epsilon0.000000
## rand steps:0
## Episode: 495 finished with steps:13, rewards:-13.000000 global step 9091
## Last 100 episodes average reward -13.000000
## Epsilon0.000000
## rand steps:0
## Episode: 496 finished with steps:13, rewards:-13.000000 global step 9104
## Last 100 episodes average reward -13.000000
## Epsilon0.000000
## rand steps:0
## Episode: 497 finished with steps:13, rewards:-13.000000 global step 9117
## Last 100 episodes average reward -13.000000
## Epsilon0.000000
## rand steps:0
## Episode: 498 finished with steps:13, rewards:-13.000000 global step 9130
## Last 100 episodes average reward -13.000000
## Epsilon0.000000
## rand steps:0
## Episode: 499 finished with steps:13, rewards:-13.000000 global step 9143
## Last 100 episodes average reward -13.000000
## Epsilon0.000000
## rand steps:0
## Episode: 500 finished with steps:13, rewards:-13.000000 global step 9156
## Last 100 episodes average reward -13.000000
## Epsilon0.000000
## rand steps:0