trainer_state.json

{
  "best_metric": 0.9085111601667893,
  "best_model_checkpoint": "layoutlmv3-test/checkpoint-1000",
  "epoch": 52.63157894736842,
  "eval_steps": 100,
  "global_step": 1000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 5.26,
      "eval_accuracy": 0.8407226910733389,
      "eval_f1": 0.8338136407300673,
      "eval_loss": 0.511518120765686,
      "eval_precision": 0.8070664807066481,
      "eval_recall": 0.862394436164928,
      "eval_runtime": 8.3121,
      "eval_samples_per_second": 6.015,
      "eval_steps_per_second": 0.842,
      "step": 100
    },
    {
      "epoch": 10.53,
      "eval_accuracy": 0.8546297396885773,
      "eval_f1": 0.8904576436222005,
      "eval_loss": 0.4660772979259491,
      "eval_precision": 0.8730310262529833,
      "eval_recall": 0.9085941381023348,
      "eval_runtime": 9.8898,
      "eval_samples_per_second": 5.056,
      "eval_steps_per_second": 0.708,
      "step": 200
    },
    {
      "epoch": 15.79,
      "eval_accuracy": 0.8552240580054677,
      "eval_f1": 0.9001475651746188,
      "eval_loss": 0.5612959861755371,
      "eval_precision": 0.8913784705309303,
      "eval_recall": 0.9090909090909091,
      "eval_runtime": 9.5301,
      "eval_samples_per_second": 5.247,
      "eval_steps_per_second": 0.735,
      "step": 300
    },
    {
      "epoch": 21.05,
      "eval_accuracy": 0.8507072387970998,
      "eval_f1": 0.8959365708622399,
      "eval_loss": 0.6766558885574341,
      "eval_precision": 0.893722194760257,
      "eval_recall": 0.8981619473422752,
      "eval_runtime": 9.557,
      "eval_samples_per_second": 5.232,
      "eval_steps_per_second": 0.732,
      "step": 400
    },
    {
      "epoch": 26.32,
      "learning_rate": 5e-06,
      "loss": 0.3022,
      "step": 500
    },
    {
      "epoch": 26.32,
      "eval_accuracy": 0.8625936051349102,
      "eval_f1": 0.9048553212359,
      "eval_loss": 0.7019667625427246,
      "eval_precision": 0.8934624697336562,
      "eval_recall": 0.9165424739195231,
      "eval_runtime": 8.8998,
      "eval_samples_per_second": 5.618,
      "eval_steps_per_second": 0.787,
      "step": 500
    },
    {
      "epoch": 31.58,
      "eval_accuracy": 0.8591465588969452,
      "eval_f1": 0.912936546974914,
      "eval_loss": 0.7107805609703064,
      "eval_precision": 0.9040428641013152,
      "eval_recall": 0.9220069547938401,
      "eval_runtime": 10.5027,
      "eval_samples_per_second": 4.761,
      "eval_steps_per_second": 0.666,
      "step": 600
    },
    {
      "epoch": 36.84,
      "eval_accuracy": 0.8516581481041245,
      "eval_f1": 0.911198815984213,
      "eval_loss": 0.7377670407295227,
      "eval_precision": 0.9049485546300833,
      "eval_recall": 0.9175360158966717,
      "eval_runtime": 8.9287,
      "eval_samples_per_second": 5.6,
      "eval_steps_per_second": 0.784,
      "step": 700
    },
    {
      "epoch": 42.11,
      "eval_accuracy": 0.8536788303815523,
      "eval_f1": 0.9117285468404228,
      "eval_loss": 0.7891532778739929,
      "eval_precision": 0.9026290165530672,
      "eval_recall": 0.9210134128166915,
      "eval_runtime": 8.67,
      "eval_samples_per_second": 5.767,
      "eval_steps_per_second": 0.807,
      "step": 800
    },
    {
      "epoch": 47.37,
      "eval_accuracy": 0.8490431475098063,
      "eval_f1": 0.909894426712497,
      "eval_loss": 0.8132860660552979,
      "eval_precision": 0.8995145631067961,
      "eval_recall": 0.9205166418281172,
      "eval_runtime": 9.3551,
      "eval_samples_per_second": 5.345,
      "eval_steps_per_second": 0.748,
      "step": 900
    },
    {
      "epoch": 52.63,
      "learning_rate": 0.0,
      "loss": 0.0223,
      "step": 1000
    },
    {
      "epoch": 52.63,
      "eval_accuracy": 0.8480922382027815,
      "eval_f1": 0.9085111601667893,
      "eval_loss": 0.8036019206047058,
      "eval_precision": 0.8972868217054264,
      "eval_recall": 0.920019870839543,
      "eval_runtime": 9.8667,
      "eval_samples_per_second": 5.068,
      "eval_steps_per_second": 0.709,
      "step": 1000
    },
    {
      "epoch": 52.63,
      "step": 1000,
      "total_flos": 2067720312311808.0,
      "train_loss": 0.16229200553894044,
      "train_runtime": 3424.968,
      "train_samples_per_second": 2.336,
      "train_steps_per_second": 0.292
    }
  ],
  "logging_steps": 500,
  "max_steps": 1000,
  "num_train_epochs": 53,
  "save_steps": 500,
  "total_flos": 2067720312311808.0,
  "trial_name": null,
  "trial_params": null
}