{
  "condition": "clean_kp",
  "config": {
    "action_dim": 7,
    "context_gain": 1.0,
    "hidden_layers": 2,
    "input_dim": 984,
    "learning_rate": 0.01,
    "mission_dim": 13,
    "momentum": 0.9,
    "reciprocal_learning_rate": 0.01,
    "weight_decay": 0.0001,
    "width": 256
  },
  "data": {
    "color_cardinality": 6,
    "elapsed_seconds": 37.730633020401,
    "env_id": "BabyAI-GoToObjS6-v1",
    "minigrid_version": "3.1.0",
    "object_cardinality": 11,
    "protocol": "babyai_shared_feedback_b0",
    "rollout_episodes": 500,
    "rollout_seed_start": 200000,
    "state_cardinality": 3,
    "train_episodes": 20000,
    "train_steps": 70159,
    "validation_episodes": 2000,
    "validation_steps": 7019,
    "vocabulary": [
      "<unk>",
      "ball",
      "blue",
      "box",
      "go",
      "green",
      "grey",
      "key",
      "purple",
      "red",
      "the",
      "to",
      "yellow"
    ]
  },
  "epoch_history": [
    {
      "epoch": 1,
      "train_loss": 0.4369402367960323
    },
    {
      "epoch": 2,
      "train_loss": 0.04315035553818399
    },
    {
      "epoch": 3,
      "train_loss": 0.018373423097655175
    },
    {
      "epoch": 4,
      "train_loss": 0.01484828113832257
    },
    {
      "epoch": 5,
      "train_loss": 0.01290607647673989
    },
    {
      "epoch": 6,
      "train_loss": 0.012369149463589895
    },
    {
      "epoch": 7,
      "train_loss": 0.014328753638441083
    },
    {
      "epoch": 8,
      "train_loss": 0.01090253034458411
    },
    {
      "epoch": 9,
      "train_loss": 0.010823998920280825
    },
    {
      "epoch": 10,
      "train_loss": 0.012246859658255496
    },
    {
      "epoch": 11,
      "train_loss": 0.011209721950687129
    },
    {
      "epoch": 12,
      "train_loss": 0.010971906608478589
    },
    {
      "epoch": 13,
      "train_loss": 0.010095493538643826
    },
    {
      "epoch": 14,
      "train_loss": 0.010141288362104784
    },
    {
      "epoch": 15,
      "train_loss": 0.010389165276640348
    },
    {
      "epoch": 16,
      "train_loss": 0.010640575006942857
    },
    {
      "epoch": 17,
      "train_loss": 0.010856473856732588
    },
    {
      "epoch": 18,
      "train_loss": 0.009704423175108704
    },
    {
      "epoch": 19,
      "train_loss": 0.009899068040011281
    },
    {
      "epoch": 20,
      "train_loss": 0.010538074031937867
    },
    {
      "epoch": 21,
      "train_loss": 0.011126897338439117
    },
    {
      "epoch": 22,
      "train_loss": 0.010471862061795864
    },
    {
      "epoch": 23,
      "train_loss": 0.00964502102123912
    },
    {
      "epoch": 24,
      "train_loss": 0.0100560691582293
    },
    {
      "epoch": 25,
      "train_loss": 0.009638783985389058
    },
    {
      "epoch": 26,
      "train_loss": 0.009922710771206766
    },
    {
      "epoch": 27,
      "train_loss": 0.00961751041320068
    },
    {
      "epoch": 28,
      "train_loss": 0.009335849603151225
    },
    {
      "epoch": 29,
      "train_loss": 0.009497242902130396
    },
    {
      "epoch": 30,
      "train_loss": 0.009709630153268914
    },
    {
      "epoch": 31,
      "train_loss": 0.009487844813530419
    },
    {
      "epoch": 32,
      "train_loss": 0.009836118470411747
    },
    {
      "epoch": 33,
      "train_loss": 0.009497633754826066
    },
    {
      "epoch": 34,
      "train_loss": 0.009304083153653086
    },
    {
      "epoch": 35,
      "train_loss": 0.009465594382596795
    },
    {
      "epoch": 36,
      "train_loss": 0.00925753563949415
    },
    {
      "epoch": 37,
      "train_loss": 0.009258489562833512
    },
    {
      "epoch": 38,
      "train_loss": 0.009257057890909809
    },
    {
      "epoch": 39,
      "train_loss": 0.00942517074491744
    },
    {
      "epoch": 40,
      "train_loss": 0.009241441347350535
    }
  ],
  "epochs_completed": 40,
  "finite": true,
  "first_nonfinite_epoch": null,
  "mission_lesion_rollout": {
    "episodes": 500,
    "mean_length": 7.192,
    "mean_return": 0.8090000000000002,
    "success": 0.888
  },
  "mission_lesion_validation": {
    "accuracy": 0.9484257016669041,
    "loss": 0.09698232739616075
  },
  "predictor": [],
  "provenance": {
    "cuda_device_name": "NVIDIA GeForce GTX 1080",
    "cuda_peak_allocated_bytes": 39800832,
    "cuda_visible_devices": "5",
    "device": "cuda:0",
    "git_commit": "e3011f144a018f6415b3c2a636c0646e412b3ced",
    "git_dirty_tracked": false,
    "minigrid_version": "3.1.0",
    "torch_version": "2.3.1+cu118"
  },
  "rollout": {
    "episodes": 500,
    "mean_length": 4.29,
    "mean_return": 0.8905500000000001,
    "success": 0.978
  },
  "stage": "babyai_shared_b1",
  "training": {
    "batch_size": 256,
    "model_seed": 4102,
    "neutral_examples_per_epoch": 0,
    "parameter_count_including_reciprocal_context_predictor": 394759,
    "shuffle_seed": 4102,
    "total_wall_seconds_including_rollouts": 185.72890639305115,
    "training_wall_seconds": 182.7532684803009
  },
  "validation": {
    "accuracy": 0.9974355321270836,
    "loss": 0.008958498200963987
  }
}