1
0
mirror of https://github.com/gryf/coach.git synced 2026-02-18 07:25:55 +01:00
Files
coach/rl_coach/traces/Atari_DQN_with_PER_pong/trace.csv
2018-08-20 13:01:30 +03:00

1.7 KiB

1Episode #Training IterIn HeatupER #TransitionsER #EpisodesEpisode LengthTotal stepsEpsilonShaped Training RewardTraining RewardUpdate Target NetworkEvaluation RewardShaped Evaluation RewardSuccess RateLoss/MeanLoss/StdevLoss/MaxLoss/MinLearning Rate/MeanLearning Rate/StdevLearning Rate/MaxLearning Rate/MinGrads (unclipped)/MeanGrads (unclipped)/StdevGrads (unclipped)/MaxGrads (unclipped)/MinQ/MeanQ/StdevQ/MaxQ/Min
210.01.01117.01117.01117.01117.01.00.0
32221.00.02002.02002.0885.02002.00.9992035000000262-21.0-21.00.00.00661135672362845460.0039462341208788630.0169418863952159883.0340672310558148e-050.00025000000000000011.0842021724855042e-190.000250.000250.0205788740.0112856080.128386970.003849274
43455.00.02938.02938.0936.02938.00.9983611000000541-20.0-20.00.00.0072206104838171910.003843868832563130.022013204172253610.00042599014705047010.00025000000000000011.0842021724855042e-190.000250.000250.0141966580000000010.00531139900000000050.0404063430.005419724599999999-0.0124267030.0214578879999999980.023741005-0.051037904
54659.00.03754.03754.0816.03754.00.997626700000078-21.0-21.00.00.0070675956867133060.003496837390859280.0167864318937063180.00049741903785616160.00025000000000000015.421010862427521e-200.000250.000250.0127322369999999990.00382579770000000040.024201730.00600734
65961.00.04961.04961.01207.04961.00.996540400000114-18.0-18.00.00.0070346627265503260.0036373643518780820.0220788903534412420.00047363861813209950.00025000000000000015.421010862427521e-200.000250.000250.0127679650.00432938150.0315004620.0063609104-0.015213630.0118595780.006441065-0.04179345