trainer_state.json

{
  "best_metric": 0.896551724137931,
  "best_model_checkpoint": "segformer-class-classWeights-augmentation/checkpoint-15",
  "epoch": 13.333333333333334,
  "global_step": 20,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.3103448275862069,
      "eval_loss": 1.0387377738952637,
      "eval_runtime": 0.934,
      "eval_samples_per_second": 31.05,
      "eval_steps_per_second": 1.071,
      "step": 1
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6551724137931034,
      "eval_loss": 1.0144158601760864,
      "eval_runtime": 1.1602,
      "eval_samples_per_second": 24.995,
      "eval_steps_per_second": 0.862,
      "step": 3
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7241379310344828,
      "eval_loss": 0.9569838047027588,
      "eval_runtime": 0.8581,
      "eval_samples_per_second": 33.797,
      "eval_steps_per_second": 1.165,
      "step": 5
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6551724137931034,
      "eval_loss": 0.9074841737747192,
      "eval_runtime": 0.8747,
      "eval_samples_per_second": 33.152,
      "eval_steps_per_second": 1.143,
      "step": 6
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.7241379310344828,
      "eval_loss": 0.8454705476760864,
      "eval_runtime": 1.0773,
      "eval_samples_per_second": 26.918,
      "eval_steps_per_second": 0.928,
      "step": 7
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.7931034482758621,
      "eval_loss": 0.7622103095054626,
      "eval_runtime": 1.2846,
      "eval_samples_per_second": 22.575,
      "eval_steps_per_second": 0.778,
      "step": 9
    },
    {
      "epoch": 6.67,
      "learning_rate": 2.777777777777778e-05,
      "loss": 0.4555,
      "step": 10
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.7931034482758621,
      "eval_loss": 0.7067170143127441,
      "eval_runtime": 0.8517,
      "eval_samples_per_second": 34.049,
      "eval_steps_per_second": 1.174,
      "step": 11
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8275862068965517,
      "eval_loss": 0.6745181083679199,
      "eval_runtime": 1.3057,
      "eval_samples_per_second": 22.211,
      "eval_steps_per_second": 0.766,
      "step": 12
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.8620689655172413,
      "eval_loss": 0.6107693314552307,
      "eval_runtime": 0.9158,
      "eval_samples_per_second": 31.668,
      "eval_steps_per_second": 1.092,
      "step": 13
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.896551724137931,
      "eval_loss": 0.5319176316261292,
      "eval_runtime": 0.8719,
      "eval_samples_per_second": 33.261,
      "eval_steps_per_second": 1.147,
      "step": 15
    },
    {
      "epoch": 11.0,
      "eval_accuracy": 0.896551724137931,
      "eval_loss": 0.49429023265838623,
      "eval_runtime": 0.8551,
      "eval_samples_per_second": 33.915,
      "eval_steps_per_second": 1.169,
      "step": 17
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.896551724137931,
      "eval_loss": 0.48956170678138733,
      "eval_runtime": 0.8629,
      "eval_samples_per_second": 33.607,
      "eval_steps_per_second": 1.159,
      "step": 18
    },
    {
      "epoch": 13.0,
      "eval_accuracy": 0.896551724137931,
      "eval_loss": 0.4820307791233063,
      "eval_runtime": 0.8493,
      "eval_samples_per_second": 34.147,
      "eval_steps_per_second": 1.177,
      "step": 19
    },
    {
      "epoch": 13.33,
      "learning_rate": 0.0,
      "loss": 0.2224,
      "step": 20
    },
    {
      "epoch": 13.33,
      "eval_accuracy": 0.896551724137931,
      "eval_loss": 0.47683313488960266,
      "eval_runtime": 1.2061,
      "eval_samples_per_second": 24.045,
      "eval_steps_per_second": 0.829,
      "step": 20
    },
    {
      "epoch": 13.33,
      "step": 20,
      "total_flos": 8.7520382681131e+16,
      "train_loss": 0.338924241065979,
      "train_runtime": 348.466,
      "train_samples_per_second": 14.98,
      "train_steps_per_second": 0.057
    }
  ],
  "max_steps": 20,
  "num_train_epochs": 20,
  "total_flos": 8.7520382681131e+16,
  "trial_name": null,
  "trial_params": null
}