{
  "train_losses": [
    -0.0008886720752343535,
    0.0018749998416751623,
    -0.0016992188757285476,
    0.0014062500558793545,
    -0.0004687501350417733,
    -2.929696347564459e-05,
    -0.0003417970729060471,
    0.00013671873603016138,
    -0.001972656464204192,
    0.0019824218470603228,
    9.765609866008162e-05,
    -0.0004492188454605639,
    -0.006416015792638063,
    0.0014746093656867743,
    -0.0011328125838190317,
    0.00041992179467342794,
    0.0007226562011055648,
    -0.0038378906901925802,
    -0.0029199221171438694,
    -0.0007714845705777407,
    0.00020507810404524207,
    0.0009277343633584678,
    0.0010449218098074198,
    -0.0011523438151925802,
    0.00015624993829987943,
    -0.000693359412252903,
    0.0012792968191206455,
    -0.001796875149011612,
    -0.0028515628073364496,
    0.00013671873603016138,
    -0.0008789062849245965,
    -0.002451172098517418,
    -0.00018554693087935448,
    -0.002089843852445483,
    -0.0010546875419095159,
    -0.002011719159781933,
    -0.0029492187313735485,
    -0.001757812686264515,
    -0.0020703126210719347,
    0.0016406249487772584,
    -0.0018164063803851604,
    -0.0008203125325962901,
    -0.0011230469681322575,
    0.0005957029643468559,
    -0.0005468751187436283,
    4.8828020226210356e-05,
    -0.0007324219914153218,
    -0.0020703128539025784,
    -0.0009960937313735485,
    0.00038085927371867,
    -2.9296934371814132e-05,
    -0.0025097657926380634,
    -0.0012402344727888703,
    -0.0011328125838190317,
    -0.003867187537252903,
    -0.0019042969215661287,
    -0.002158203162252903,
    -4.8828194849193096e-05,
    0.0011621094308793545,
    -0.002871094038709998,
    0.0007421874906867743,
    -0.001679687644354999,
    3.9062462747097015e-05,
    -0.0018652346916496754,
    -0.006669921800494194,
    -0.0012500000884756446,
    0.00041015614988282323,
    -0.0017382815713062882,
    -1.9531260477378964e-05,
    -0.0010839845053851604,
    -0.000820312590803951,
    -0.0018652344588190317,
    -0.001064453274011612,
    -0.0021093753166496754,
    -0.002910156501457095,
    0.0013671874767169356,
    6.835930980741978e-05,
    -0.0012109376257285476,
    -0.0013281251303851604,
    0.0001074218307621777,
    -0.002001953311264515,
    -0.0029980470426380634,
    3.9062462747097015e-05,
    1.9531173165887594e-05,
    -0.0007617189548909664,
    -0.0020996094681322575,
    -0.0006542970659211278,
    -0.0020703128539025784,
    0.0009179686894640326,
    -0.0018457032274454832,
    -0.0014550783671438694,
    3.9062375435605645e-05,
    -0.002050781389698386,
    -0.0021289063151925802,
    -0.0016503906808793545,
    -0.0003515625139698386,
    -0.00020507819135673344,
    -0.0005859376396983862,
    -0.001015624962747097,
    -0.0008007813594304025,
    0.0001269530621357262,
    -0.0007421876071020961,
    -0.0030859375838190317,
    0.0011132812360301614,
    -0.0017675783019512892,
    -4.8828194849193096e-05,
    0.0003710936871357262,
    -0.0012500000884756446,
    -0.003554687835276127,
    -0.0015332032926380634,
    -0.0014941407134756446,
    -0.0010351561941206455,
    -0.001601562718860805,
    -0.0005175783298909664,
    0.0013281248975545168,
    0.001015624962747097,
    -0.0021289063151925802,
    -0.0005371093284338713,
    0.0015917968703433871,
    -0.0018164064968004823,
    -0.0019628906156867743,
    -0.002080078236758709,
    -4.882816574536264e-05,
    -0.0002441406832076609,
    -0.0024609374813735485,
    0.0024218750186264515,
    -0.0006054689292795956,
    -0.0012109377421438694,
    0.0010253905784338713,
    -0.0001269530621357262,
    0.0017285156063735485,
    0.0012890624348074198,
    0.001376952975988388,
    0.0008789062267169356,
    0.0013671874767169356,
    -0.0024023440200835466,
    -0.0009570313850417733,
    0.0014746092492714524,
    -0.003847656538709998,
    -0.0036914064548909664,
    0.00017578122788108885,
    -0.0011132813524454832,
    -0.00025390629889443517,
    7.812489639036357e-05,
    -0.0019433595007285476,
    -0.0008007813012227416,
    -0.00278320349752903,
    0.0002636718563735485,
    0.002041015774011612,
    -0.0008300782646983862,
    -0.0005566407926380634,
    0.001943359267897904,
    -0.0008789063431322575,
    -0.0001562500256113708,
    -0.0032324218191206455,
    5.859369412064552e-05,
    4.882807843387127e-05,
    -0.0002636719145812094,
    9.765612776391208e-05,
    -0.00042968758498318493,
    -0.0018945314222946763,
    -0.002119140699505806,
    -0.0029882811941206455,
    -6.835945532657206e-05,
    0.0011230468517169356,
    -0.00020507816225290298,
    -0.0013183593982830644,
    -0.0011230469681322575,
    -0.0008984374580904841,
    -0.001972656464204192,
    0.00025390618247911334,
    0.0011035155039280653,
    -0.0020605470053851604,
    -0.0017773439176380634,
    0.0002441405667923391,
    -0.00498046912252903,
    0.0009570312686264515,
    -0.0021484375465661287,
    -0.0019433596171438694,
    0.0002441405667923391,
    0.0011816405458375812,
    0.00019531240104697645,
    -0.0010058595798909664,
    -0.0022753907833248377,
    -0.0019824218470603228,
    0.0008203124161809683,
    -0.0008105469751171768,
    -0.0014160156715661287,
    -0.0009082031901925802,
    0.0008886718424037099,
    0.0009863279992714524,
    -2.9296992579475045e-05,
    -0.003017578274011612,
    -0.0029199218843132257,
    -0.0020117186941206455
  ],
  "test_losses": [
    -0.1667999999999999,
    -0.014999999999999871,
    -0.12039999999999987,
    -0.0959999999999999,
    -0.1423999999999999,
    -0.32739999999999986,
    -0.2145999999999999,
    -0.3208,
    -0.24459999999999993,
    -0.21639999999999993,
    -0.049999999999999815,
    -0.07439999999999984,
    -0.12819999999999993,
    -0.10779999999999988,
    -0.08139999999999988,
    -0.13039999999999993,
    -0.1643999999999999,
    -0.2215999999999999,
    -0.2357999999999999
  ],
  "config": {
    "training_type": "online_ppo",
    "environment": "MiniHack-Room-Random-15x15-v0",
    "total_timesteps": 195,
    "training_time": 4420.289131402969,
    "device": "cuda",
    "ppo_config": {
      "learning_rate": 0.0003,
      "n_epochs": 4,
      "gamma": 0.99,
      "vf_coef": 0.5,
      "ent_coef": 0.01,
      "max_grad_norm": 0.5
    },
    "exploration_config": {
      "use_curiosity": true,
      "curiosity_dyn": true,
      "curiosity_skill_entropy": true,
      "curiosity_skill_transition_novelty": true,
      "curiosity_dyn_coef": 0.03,
      "curiosity_hdp_coef": 0.002,
      "curiosity_stn_coef": 0.0005,
      "use_rnd": false,
      "rnd_lr": null,
      "rnd_coef": null
    },
    "model_sources": {
      "vae_repo_id": "CatkinChen/nethack-vae-hmm",
      "hmm_repo_id": "CatkinChen/nethack-hmm"
    }
  },
  "final_train_loss": -0.0020117186941206455,
  "final_test_loss": -0.2357999999999999,
  "total_epochs": 195,
  "best_train_loss": -0.006669921800494194,
  "best_test_loss": -0.32739999999999986
}