checkpoint-81/trainer_state.json

{
  "best_metric": 0.2619660496711731,
  "best_model_checkpoint": "longformer-spans/checkpoint-81",
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 81,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_B": {
        "f1-score": 0.816,
        "precision": 0.7461594732991953,
        "recall": 0.9002647837599294,
        "support": 1133.0
      },
      "eval_I": {
        "f1-score": 0.9321376763813793,
        "precision": 0.9024103768767235,
        "recall": 0.9638902525500463,
        "support": 18333.0
      },
      "eval_O": {
        "f1-score": 0.8560784528570645,
        "precision": 0.931782945736434,
        "recall": 0.7917511147142278,
        "support": 9868.0
      },
      "eval_accuracy": 0.9035249198881844,
      "eval_loss": 0.2619660496711731,
      "eval_macro avg": {
        "f1-score": 0.8680720430794812,
        "precision": 0.860117598637451,
        "recall": 0.8853020503414012,
        "support": 29334.0
      },
      "eval_runtime": 5.5032,
      "eval_samples_per_second": 14.537,
      "eval_steps_per_second": 1.817,
      "eval_weighted avg": {
        "f1-score": 0.9020655278480035,
        "precision": 0.9062562975065145,
        "recall": 0.9035249198881844,
        "support": 29334.0
      },
      "step": 81
    }
  ],
  "logging_steps": 500,
  "max_steps": 1620,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 20,
  "save_steps": 500,
  "total_flos": 143786652735600.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}