{
  "config": {
    "epochs": 50,
    "lr": 0.0001,
    "warmup_steps": 100,
    "grad_accum_steps": 8,
    "grad_clip": 1.0
  },
  "train": [
    {
      "epoch": 1,
      "avg_loss": 48.950583,
      "lr": 9.99783297632528e-05,
      "step": 192,
      "time_s": 1012.5
    },
    {
      "epoch": 2,
      "avg_loss": 39.822198,
      "lr": 9.979363170168663e-05,
      "step": 384,
      "time_s": 1016.4
    },
    {
      "epoch": 3,
      "avg_loss": 36.55349,
      "lr": 9.942103713803461e-05,
      "step": 576,
      "time_s": 1018.1
    },
    {
      "epoch": 4,
      "avg_loss": 34.752199,
      "lr": 9.88620263818582e-05,
      "step": 768,
      "time_s": 1016.3
    },
    {
      "epoch": 5,
      "avg_loss": 33.489609,
      "lr": 9.811882036992912e-05,
      "step": 960,
      "time_s": 1016.9
    }
  ],
  "val": [
    {
      "epoch": 1,
      "avg_loss": 42.255684
    },
    {
      "epoch": 2,
      "avg_loss": 37.861168
    },
    {
      "epoch": 3,
      "avg_loss": 35.591921
    },
    {
      "epoch": 4,
      "avg_loss": 34.174768
    },
    {
      "epoch": 5,
      "avg_loss": 33.054291
    }
  ],
  "sanity": [
    {
      "epoch": 5,
      "audio_path": "/kaggle/working/inference_out.wav"
    }
  ]
}