{
  "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": 0.35939537301659585
    },
    {
      "epoch": 2,
      "train_loss": 0.04168066847730767
    },
    {
      "epoch": 3,
      "train_loss": 0.01859026498753916
    },
    {
      "epoch": 4,
      "train_loss": 0.015395146521993659
    },
    {
      "epoch": 5,
      "train_loss": 0.015227786766534503
    },
    {
      "epoch": 6,
      "train_loss": 0.012962078500369734
    },
    {
      "epoch": 7,
      "train_loss": 0.013404185661241751
    },
    {
      "epoch": 8,
      "train_loss": 0.01239592205690728
    },
    {
      "epoch": 9,
      "train_loss": 0.012198219955157996
    },
    {
      "epoch": 10,
      "train_loss": 0.011407826012000442
    },
    {
      "epoch": 11,
      "train_loss": 0.010740509365939281
    },
    {
      "epoch": 12,
      "train_loss": 0.01174285890576853
    },
    {
      "epoch": 13,
      "train_loss": 0.012136201211251318
    },
    {
      "epoch": 14,
      "train_loss": 0.010752951244569637
    },
    {
      "epoch": 15,
      "train_loss": 0.010369447311809794
    },
    {
      "epoch": 16,
      "train_loss": 0.01007143171355975
    },
    {
      "epoch": 17,
      "train_loss": 0.011025185833727432
    },
    {
      "epoch": 18,
      "train_loss": 0.010670005309564823
    },
    {
      "epoch": 19,
      "train_loss": 0.010151156608692624
    },
    {
      "epoch": 20,
      "train_loss": 0.010251642129891976
    },
    {
      "epoch": 21,
      "train_loss": 0.01011152534618635
    },
    {
      "epoch": 22,
      "train_loss": 0.009685466071823612
    },
    {
      "epoch": 23,
      "train_loss": 0.01069813368905505
    },
    {
      "epoch": 24,
      "train_loss": 0.010710137346018615
    },
    {
      "epoch": 25,
      "train_loss": 0.01011071698105132
    },
    {
      "epoch": 26,
      "train_loss": 0.009644118504260075
    },
    {
      "epoch": 27,
      "train_loss": 0.009702050907283344
    },
    {
      "epoch": 28,
      "train_loss": 0.010063305073708761
    },
    {
      "epoch": 29,
      "train_loss": 0.009666770600332794
    },
    {
      "epoch": 30,
      "train_loss": 0.009591050190554763
    },
    {
      "epoch": 31,
      "train_loss": 0.00987653798296709
    },
    {
      "epoch": 32,
      "train_loss": 0.00944871237702583
    },
    {
      "epoch": 33,
      "train_loss": 0.009831787788584321
    },
    {
      "epoch": 34,
      "train_loss": 0.009708568485220894
    },
    {
      "epoch": 35,
      "train_loss": 0.00941432591813447
    },
    {
      "epoch": 36,
      "train_loss": 0.009692842404315756
    },
    {
      "epoch": 37,
      "train_loss": 0.009456626759545709
    },
    {
      "epoch": 38,
      "train_loss": 0.009676380681551315
    },
    {
      "epoch": 39,
      "train_loss": 0.009368902054848149
    },
    {
      "epoch": 40,
      "train_loss": 0.009439258407060565
    }
  ],
  "epochs_completed": 40,
  "finite": true,
  "first_nonfinite_epoch": null,
  "mission_lesion_rollout": {
    "episodes": 500,
    "mean_length": 12.65,
    "mean_return": 0.65555,
    "success": 0.718
  },
  "mission_lesion_validation": {
    "accuracy": 0.8236215985183074,
    "loss": 0.5513675970677246
  },
  "predictor": [],
  "provenance": {
    "cuda_device_name": "NVIDIA GeForce GTX 1080",
    "cuda_peak_allocated_bytes": 39800832,
    "cuda_visible_devices": "0",
    "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": 4101,
    "neutral_examples_per_epoch": 0,
    "parameter_count_including_reciprocal_context_predictor": 394759,
    "shuffle_seed": 4101,
    "total_wall_seconds_including_rollouts": 184.13239169120789,
    "training_wall_seconds": 178.87876749038696
  },
  "validation": {
    "accuracy": 0.9974355321270836,
    "loss": 0.007173381226747901
  }
}