trainer_state.json

{
  "best_metric": 0.896551724137931,
  "best_model_checkpoint": "segformer-class-classWeights-augmentation/checkpoint-13",
  "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": 0.27586206896551724,
      "eval_loss": 1.1699918508529663,
      "eval_runtime": 1.1929,
      "eval_samples_per_second": 24.31,
      "eval_steps_per_second": 0.838,
      "step": 1
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.3793103448275862,
      "eval_loss": 1.0350514650344849,
      "eval_runtime": 0.7976,
      "eval_samples_per_second": 36.36,
      "eval_steps_per_second": 1.254,
      "step": 3
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.5172413793103449,
      "eval_loss": 0.9730662107467651,
      "eval_runtime": 0.8014,
      "eval_samples_per_second": 36.186,
      "eval_steps_per_second": 1.248,
      "step": 5
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.5172413793103449,
      "eval_loss": 0.9361577033996582,
      "eval_runtime": 1.0363,
      "eval_samples_per_second": 27.984,
      "eval_steps_per_second": 0.965,
      "step": 6
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.5517241379310345,
      "eval_loss": 0.8889826536178589,
      "eval_runtime": 0.7918,
      "eval_samples_per_second": 36.624,
      "eval_steps_per_second": 1.263,
      "step": 7
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.7586206896551724,
      "eval_loss": 0.7596346735954285,
      "eval_runtime": 0.8269,
      "eval_samples_per_second": 35.071,
      "eval_steps_per_second": 1.209,
      "step": 9
    },
    {
      "epoch": 6.67,
      "learning_rate": 2.777777777777778e-05,
      "loss": 0.5024,
      "step": 10
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8620689655172413,
      "eval_loss": 0.6531320810317993,
      "eval_runtime": 0.7953,
      "eval_samples_per_second": 36.466,
      "eval_steps_per_second": 1.257,
      "step": 11
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8620689655172413,
      "eval_loss": 0.6170477271080017,
      "eval_runtime": 1.0429,
      "eval_samples_per_second": 27.806,
      "eval_steps_per_second": 0.959,
      "step": 12
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.896551724137931,
      "eval_loss": 0.5878463387489319,
      "eval_runtime": 0.7888,
      "eval_samples_per_second": 36.767,
      "eval_steps_per_second": 1.268,
      "step": 13
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.8620689655172413,
      "eval_loss": 0.5418170094490051,
      "eval_runtime": 0.8158,
      "eval_samples_per_second": 35.546,
      "eval_steps_per_second": 1.226,
      "step": 15
    },
    {
      "epoch": 11.0,
      "eval_accuracy": 0.8620689655172413,
      "eval_loss": 0.5121886730194092,
      "eval_runtime": 0.8148,
      "eval_samples_per_second": 35.592,
      "eval_steps_per_second": 1.227,
      "step": 17
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.8620689655172413,
      "eval_loss": 0.5020919442176819,
      "eval_runtime": 1.0894,
      "eval_samples_per_second": 26.62,
      "eval_steps_per_second": 0.918,
      "step": 18
    },
    {
      "epoch": 13.0,
      "eval_accuracy": 0.8620689655172413,
      "eval_loss": 0.4927540421485901,
      "eval_runtime": 0.8053,
      "eval_samples_per_second": 36.014,
      "eval_steps_per_second": 1.242,
      "step": 19
    },
    {
      "epoch": 13.33,
      "learning_rate": 0.0,
      "loss": 0.3117,
      "step": 20
    },
    {
      "epoch": 13.33,
      "eval_accuracy": 0.8620689655172413,
      "eval_loss": 0.4871620237827301,
      "eval_runtime": 0.7941,
      "eval_samples_per_second": 36.521,
      "eval_steps_per_second": 1.259,
      "step": 20
    },
    {
      "epoch": 13.33,
      "step": 20,
      "total_flos": 8.7520382681131e+16,
      "train_loss": 0.4070509672164917,
      "train_runtime": 340.2477,
      "train_samples_per_second": 15.342,
      "train_steps_per_second": 0.059
    }
  ],
  "max_steps": 20,
  "num_train_epochs": 20,
  "total_flos": 8.7520382681131e+16,
  "trial_name": null,
  "trial_params": null
}