{
  "config": {
    "total_timesteps": 500000,
    "lr": 0.00025,
    "batch_size": 128,
    "n_envs": 2,
    "max_clicks": 5,
    "sam_model": "facebook/sam2.1-hiera-base-plus",
    "obs_size": 128,
    "grid_size": 32,
    "features_dim": 512,
    "clip_range": 0.1,
    "ent_coef": 0.02
  },
  "training_time_minutes": 185.7014569679896,
  "oracle_baseline": {
    "step_0": {
      "mean_dice": 0.0,
      "std_dice": 0.0,
      "median_dice": 0.0,
      "n_samples": 100
    },
    "step_1": {
      "mean_dice": 0.7482452005228067,
      "std_dice": 0.31596950664171314,
      "median_dice": 0.9178391060211384,
      "n_samples": 100
    },
    "step_2": {
      "mean_dice": 0.8480271216874984,
      "std_dice": 0.21416851137760265,
      "median_dice": 0.9394086188508548,
      "n_samples": 100
    },
    "step_3": {
      "mean_dice": 0.8545456562888544,
      "std_dice": 0.2173320841160492,
      "median_dice": 0.947919676669708,
      "n_samples": 100
    },
    "step_4": {
      "mean_dice": 0.8942226411519251,
      "std_dice": 0.16921504649215535,
      "median_dice": 0.9531381876940285,
      "n_samples": 100
    },
    "step_5": {
      "mean_dice": 0.9170403233154673,
      "std_dice": 0.12807091406498944,
      "median_dice": 0.9576795517312017,
      "n_samples": 100
    }
  },
  "rl_agent": {
    "step_0": {
      "mean_dice": 0.7482452005228069,
      "std_dice": 0.31596950664171314,
      "median_dice": 0.9178391060211384,
      "n_samples": 100
    },
    "step_1": {
      "mean_dice": 0.652811772161106,
      "std_dice": 0.3374588502417598,
      "median_dice": 0.8138924606580109,
      "n_samples": 100
    },
    "step_2": {
      "mean_dice": 0.624163164298948,
      "std_dice": 0.35094013693186865,
      "median_dice": 0.8076231453905804,
      "n_samples": 100
    },
    "step_3": {
      "mean_dice": 0.6667150707807159,
      "std_dice": 0.32462124639427187,
      "median_dice": 0.8261332535840864,
      "n_samples": 100
    },
    "step_4": {
      "mean_dice": 0.6444545406611362,
      "std_dice": 0.30873485413441143,
      "median_dice": 0.7364939532431213,
      "n_samples": 100
    },
    "step_5": {
      "mean_dice": 0.6140601785851652,
      "std_dice": 0.32327465328286625,
      "median_dice": 0.7070430070530784,
      "n_samples": 100
    },
    "mean_episode_reward": -0.03481383275250369
  },
  "total_params": 1886625
}