trainer_state.json

{
  "best_global_step": 144,
  "best_metric": 0.7762978142076503,
  "best_model_checkpoint": "./results/PlanTL-GOB-ES/roberta-base-ca_fold_2/checkpoint-144",
  "epoch": 7.0,
  "eval_steps": 500,
  "global_step": 252,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7916666666666666,
      "eval_f1": 0.6996124031007752,
      "eval_loss": 1.1276955604553223,
      "eval_precision": 0.6267361111111112,
      "eval_recall": 0.7916666666666666,
      "eval_runtime": 0.2168,
      "eval_samples_per_second": 332.118,
      "eval_steps_per_second": 41.515,
      "step": 36
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7916666666666666,
      "eval_f1": 0.6996124031007752,
      "eval_loss": 1.0436447858810425,
      "eval_precision": 0.6267361111111112,
      "eval_recall": 0.7916666666666666,
      "eval_runtime": 0.2157,
      "eval_samples_per_second": 333.721,
      "eval_steps_per_second": 41.715,
      "step": 72
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7916666666666666,
      "eval_f1": 0.73,
      "eval_loss": 1.036052942276001,
      "eval_precision": 0.6819852941176471,
      "eval_recall": 0.7916666666666666,
      "eval_runtime": 0.2154,
      "eval_samples_per_second": 334.325,
      "eval_steps_per_second": 41.791,
      "step": 108
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8194444444444444,
      "eval_f1": 0.7762978142076503,
      "eval_loss": 1.0554141998291016,
      "eval_precision": 0.7532051282051282,
      "eval_recall": 0.8194444444444444,
      "eval_runtime": 0.2164,
      "eval_samples_per_second": 332.729,
      "eval_steps_per_second": 41.591,
      "step": 144
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8194444444444444,
      "eval_f1": 0.7695185185185186,
      "eval_loss": 1.1022535562515259,
      "eval_precision": 0.7769607843137254,
      "eval_recall": 0.8194444444444444,
      "eval_runtime": 0.2113,
      "eval_samples_per_second": 340.686,
      "eval_steps_per_second": 42.586,
      "step": 180
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.7777777777777778,
      "eval_f1": 0.7620167189132706,
      "eval_loss": 1.2096033096313477,
      "eval_precision": 0.7472928436911488,
      "eval_recall": 0.7777777777777778,
      "eval_runtime": 0.2124,
      "eval_samples_per_second": 338.991,
      "eval_steps_per_second": 42.374,
      "step": 216
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8055555555555556,
      "eval_f1": 0.7725789313904068,
      "eval_loss": 1.1715612411499023,
      "eval_precision": 0.7595441595441597,
      "eval_recall": 0.8055555555555556,
      "eval_runtime": 0.2136,
      "eval_samples_per_second": 337.067,
      "eval_steps_per_second": 42.133,
      "step": 252
    }
  ],
  "logging_steps": 500,
  "max_steps": 720,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 20,
  "save_steps": 500,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 3,
        "early_stopping_threshold": 0.0
      },
      "attributes": {
        "early_stopping_patience_counter": 3
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 42183669924396.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}