trainer_state.json

{
  "best_global_step": 108,
  "best_metric": 0.7902222222222223,
  "best_model_checkpoint": "./results/PlanTL-GOB-ES/roberta-base-ca_fold_0/checkpoint-108",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 216,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8055555555555556,
      "eval_f1": 0.7188034188034188,
      "eval_loss": 1.106392502784729,
      "eval_precision": 0.6489197530864197,
      "eval_recall": 0.8055555555555556,
      "eval_runtime": 0.2304,
      "eval_samples_per_second": 312.547,
      "eval_steps_per_second": 39.068,
      "step": 36
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8055555555555556,
      "eval_f1": 0.7188034188034188,
      "eval_loss": 1.0426812171936035,
      "eval_precision": 0.6489197530864197,
      "eval_recall": 0.8055555555555556,
      "eval_runtime": 0.2277,
      "eval_samples_per_second": 316.231,
      "eval_steps_per_second": 39.529,
      "step": 72
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8472222222222222,
      "eval_f1": 0.7902222222222223,
      "eval_loss": 0.9810686111450195,
      "eval_precision": 0.7408789386401327,
      "eval_recall": 0.8472222222222222,
      "eval_runtime": 0.2322,
      "eval_samples_per_second": 310.083,
      "eval_steps_per_second": 38.76,
      "step": 108
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8055555555555556,
      "eval_f1": 0.7636198046034112,
      "eval_loss": 0.9648534059524536,
      "eval_precision": 0.7259837962962963,
      "eval_recall": 0.8055555555555556,
      "eval_runtime": 0.221,
      "eval_samples_per_second": 325.728,
      "eval_steps_per_second": 40.716,
      "step": 144
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.7916666666666666,
      "eval_f1": 0.7743751070022257,
      "eval_loss": 1.111234188079834,
      "eval_precision": 0.7583333333333333,
      "eval_recall": 0.7916666666666666,
      "eval_runtime": 0.219,
      "eval_samples_per_second": 328.811,
      "eval_steps_per_second": 41.101,
      "step": 180
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8333333333333334,
      "eval_f1": 0.7781036168132941,
      "eval_loss": 1.070778727531433,
      "eval_precision": 0.7297979797979798,
      "eval_recall": 0.8333333333333334,
      "eval_runtime": 0.2331,
      "eval_samples_per_second": 308.844,
      "eval_steps_per_second": 38.606,
      "step": 216
    }
  ],
  "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": 32629877084376.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}