training_metrics.json

{
  "task_id": "easy_screening",
  "n_episodes": 30,
  "hyperparameters": {
    "lr": 0.0003,
    "gamma": 0.99,
    "entropy_coeff": 0.02,
    "batch_size": 5,
    "hidden_dim": 128,
    "state_dim": 16,
    "action_dim": 166
  },
  "episode_rewards": [
    0.47,
    1.1073118279569893,
    1.1486231884057971,
    1.1336231884057972,
    0.405,
    0.395,
    1.1296806853582555,
    0.38,
    0.7283823529411766,
    0.0,
    0.39,
    -0.095,
    1.137962962962963,
    1.1951785714285714,
    0.9053636363636364,
    -0.01754088050314473,
    -0.04,
    -0.06,
    0.0,
    0.22666666666666668,
    0.435,
    0.45,
    0.45,
    0.37666666666666665,
    0.435,
    0.455,
    0.5412162162162163,
    0.33899999999999997,
    0.3416666666666666,
    0.42
  ],
  "episode_grader_scores": [
    0.5,
    0.8306451612903226,
    0.8369565217391305,
    0.8369565217391305,
    0.5,
    0.5,
    0.9688473520249221,
    0.5,
    0.7058823529411765,
    0.0,
    0.5,
    0.0,
    0.837962962962963,
    0.9776785714285714,
    0.8863636363636364,
    0.053459119496855306,
    0.0,
    0.0,
    0.0,
    0.5,
    0.5,
    0.5,
    0.5,
    0.5,
    0.5,
    0.5,
    0.5495495495495496,
    0.5,
    0.27499999999999997,
    0.5
  ],
  "episode_steps": [
    5,
    4,
    3,
    4,
    6,
    6,
    5,
    7,
    4,
    1,
    7,
    8,
    4,
    3,
    10,
    6,
    3,
    7,
    1,
    4,
    4,
    3,
    3,
    5,
    4,
    6,
    4,
    7,
    2,
    5
  ],
  "policy_losses": [
    0.28967130184173584,
    0.05730011314153671,
    -0.06924888491630554,
    -0.28697478771209717,
    -0.1783256083726883,
    -0.12063005566596985
  ],
  "value_losses": [
    0.39626142382621765,
    0.24146510660648346,
    0.29013994336128235,
    0.06388193368911743,
    0.02375689707696438,
    0.02241377718746662
  ],
  "best_avg_grader_score": 0.6179287909734683,
  "total_training_time_s": 0.13345718383789062,
  "eval_stochastic_avg_reward": 0.5792066304974347,
  "eval_stochastic_avg_grader_score": 0.5784816304974348,
  "eval_stochastic_avg_steps": 5.6,
  "eval_stochastic_rewards": [
    1.0402956989247312,
    0.485,
    -0.11333333333333336,
    0.455,
    0.36250000000000004,
    1.0112089552238808,
    0.41,
    0.21499999999999997,
    1.1173118279569894,
    0.883621495327103,
    0.45,
    1.1073118279569893,
    0.8680882352941177,
    0.405,
    0.675031128404669,
    0.45,
    -0.11,
    0.415,
    0.32,
    1.1370967741935485
  ],
  "eval_stochastic_grader_scores": [
    0.8494623655913979,
    0.5,
    0.0,
    0.5,
    0.5,
    0.9207089552238806,
    0.5,
    0.5,
    0.8306451612903226,
    0.8411214953271029,
    0.5,
    0.8306451612903226,
    0.803921568627451,
    0.5,
    0.6060311284046691,
    0.5,
    0.0,
    0.5,
    0.5,
    0.8870967741935485
  ],
  "eval_greedy_avg_reward": 0.3627500000000001,
  "eval_greedy_avg_grader_score": 0.425,
  "eval_greedy_avg_steps": 6.45,
  "eval_greedy_rewards": [
    0.44,
    0.455,
    -0.08,
    0.455,
    -0.06,
    0.39,
    0.455,
    0.455,
    0.455,
    -0.06,
    0.42,
    0.455,
    0.41000000000000003,
    0.455,
    0.44,
    0.39,
    0.41000000000000003,
    0.49,
    0.44,
    0.44
  ],
  "eval_greedy_grader_scores": [
    0.5,
    0.5,
    0.0,
    0.5,
    0.0,
    0.5,
    0.5,
    0.5,
    0.5,
    0.0,
    0.5,
    0.5,
    0.5,
    0.5,
    0.5,
    0.5,
    0.5,
    0.5,
    0.5,
    0.5
  ]
}