{
  "condition": "bp",
  "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": 0.35355447364124387
    },
    {
      "epoch": 2,
      "train_loss": 0.042185719047080385
    },
    {
      "epoch": 3,
      "train_loss": 0.019281441472809423
    },
    {
      "epoch": 4,
      "train_loss": 0.015513235063884747
    },
    {
      "epoch": 5,
      "train_loss": 0.013296923749995502
    },
    {
      "epoch": 6,
      "train_loss": 0.012671216844411736
    },
    {
      "epoch": 7,
      "train_loss": 0.014801131509646605
    },
    {
      "epoch": 8,
      "train_loss": 0.011190175573574379
    },
    {
      "epoch": 9,
      "train_loss": 0.011067502123346045
    },
    {
      "epoch": 10,
      "train_loss": 0.012195465334860439
    },
    {
      "epoch": 11,
      "train_loss": 0.011578771525104954
    },
    {
      "epoch": 12,
      "train_loss": 0.011191591258076105
    },
    {
      "epoch": 13,
      "train_loss": 0.010304119926665656
    },
    {
      "epoch": 14,
      "train_loss": 0.010280991342045706
    },
    {
      "epoch": 15,
      "train_loss": 0.010555616432949054
    },
    {
      "epoch": 16,
      "train_loss": 0.010838844217995013
    },
    {
      "epoch": 17,
      "train_loss": 0.011003393904221328
    },
    {
      "epoch": 18,
      "train_loss": 0.00977229246150025
    },
    {
      "epoch": 19,
      "train_loss": 0.010014041190416637
    },
    {
      "epoch": 20,
      "train_loss": 0.010900152776686644
    },
    {
      "epoch": 21,
      "train_loss": 0.011283926658798009
    },
    {
      "epoch": 22,
      "train_loss": 0.010686473336536438
    },
    {
      "epoch": 23,
      "train_loss": 0.009769557828972625
    },
    {
      "epoch": 24,
      "train_loss": 0.010240042670841583
    },
    {
      "epoch": 25,
      "train_loss": 0.009798416777471588
    },
    {
      "epoch": 26,
      "train_loss": 0.01003834310199388
    },
    {
      "epoch": 27,
      "train_loss": 0.00971504575424743
    },
    {
      "epoch": 28,
      "train_loss": 0.009448815450199287
    },
    {
      "epoch": 29,
      "train_loss": 0.009595394653098827
    },
    {
      "epoch": 30,
      "train_loss": 0.009837813718159769
    },
    {
      "epoch": 31,
      "train_loss": 0.009620558681940151
    },
    {
      "epoch": 32,
      "train_loss": 0.009997257824500345
    },
    {
      "epoch": 33,
      "train_loss": 0.009587076458529653
    },
    {
      "epoch": 34,
      "train_loss": 0.009427764000191184
    },
    {
      "epoch": 35,
      "train_loss": 0.009570616646458141
    },
    {
      "epoch": 36,
      "train_loss": 0.009372298921966418
    },
    {
      "epoch": 37,
      "train_loss": 0.009358159794302826
    },
    {
      "epoch": 38,
      "train_loss": 0.00933290207785004
    },
    {
      "epoch": 39,
      "train_loss": 0.009472835550338708
    },
    {
      "epoch": 40,
      "train_loss": 0.009313055751527743
    }
  ],
  "epochs_completed": 40,
  "finite": true,
  "first_nonfinite_epoch": null,
  "mission_lesion_rollout": {
    "episodes": 500,
    "mean_length": 4.61,
    "mean_return": 0.8815500000000002,
    "success": 0.968
  },
  "mission_lesion_validation": {
    "accuracy": 0.9920216555064824,
    "loss": 0.0308635174710113
  },
  "predictor": [],
  "provenance": {
    "cuda_device_name": "NVIDIA GeForce GTX 1080",
    "cuda_peak_allocated_bytes": 39800832,
    "cuda_visible_devices": "4",
    "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": 180.92169880867004,
    "training_wall_seconds": 176.77257347106934
  },
  "validation": {
    "accuracy": 0.9974355321270836,
    "loss": 0.009139203621054123
  }
}