1
0
mirror of https://github.com/gryf/coach.git synced 2026-01-04 04:44:15 +01:00
Files
coach/rl_coach/traces/Atari_Bootstrapped_DQN_pong/trace.csv
2018-10-02 17:55:16 +03:00

1.9 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)/MinDiscounted Return/MeanDiscounted Return/StdevDiscounted Return/MaxDiscounted Return/MinQ/MeanQ/StdevQ/MaxQ/Min
210.01.0986.0986.0986.0986.07.00.0-1.82055450768214190.7192845707051421-0.2081522550905921-3.1698994392478896
320.01.01806.01806.0820.01806.04.00.0-2.33709693943518640.575288014748253-0.7105532272722921-3.355172823288848
43206.00.02629.02629.0823.02629.05.0-21.0-21.00.00.0141866319051048560.0136553082002718280.069096945226192470.00054609170183539380.00025000000000000011.0842021724855042e-190.000250.000250.0149388180000000010.00552471870.0347803570000000050.0049935523-2.33427228363145020.7834970909114538-0.38878391807422696-3.369599601005491
54398.00.03397.03397.0768.03397.03.0-21.0-21.00.00.0145184190235643960.0132562144750883860.064406834542751310.00059352372772991660.00025000000000000015.421010862427521e-200.000250.000250.0136187520.0038833050.0283201360.0057370984-2.44951404116649260.5558315778011723-0.7105532272722921-3.354852824180864
65705.00.04626.04626.01229.04626.06.0-19.0-19.00.00.0139123145572413420.0135732583275542680.080492578446865080.000383269827580079330.00025000000000000015.421010862427521e-200.000250.000250.01296384350.0048549210.0359246660.0042663114-1.44694280475364030.7634920719307412-0.008604775224526406-3.170625540860168-0.0139955090.0129839839999999990.019298933-0.037532326