trainer_state.json

{
  "best_metric": 0.9077802634849614,
  "best_model_checkpoint": "layoutlmv3-test/checkpoint-1000",
  "epoch": 100.0,
  "global_step": 1000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 10.0,
      "eval_accuracy": 0.8409508497932935,
      "eval_f1": 0.8606119475473532,
      "eval_loss": 0.5238379240036011,
      "eval_precision": 0.8366383380547686,
      "eval_recall": 0.886,
      "eval_runtime": 4.4875,
      "eval_samples_per_second": 12.033,
      "eval_steps_per_second": 0.891,
      "step": 100
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 0.8322232429949472,
      "eval_f1": 0.8856507779698691,
      "eval_loss": 0.6930040717124939,
      "eval_precision": 0.8750610053684724,
      "eval_recall": 0.8965,
      "eval_runtime": 4.6295,
      "eval_samples_per_second": 11.664,
      "eval_steps_per_second": 0.864,
      "step": 200
    },
    {
      "epoch": 30.0,
      "eval_accuracy": 0.8414101975195223,
      "eval_f1": 0.899009900990099,
      "eval_loss": 0.7784350514411926,
      "eval_precision": 0.8901960784313725,
      "eval_recall": 0.908,
      "eval_runtime": 4.5256,
      "eval_samples_per_second": 11.932,
      "eval_steps_per_second": 0.884,
      "step": 300
    },
    {
      "epoch": 40.0,
      "eval_accuracy": 0.836357372531006,
      "eval_f1": 0.8982630272952853,
      "eval_loss": 0.9056066870689392,
      "eval_precision": 0.8916256157635468,
      "eval_recall": 0.905,
      "eval_runtime": 4.5498,
      "eval_samples_per_second": 11.869,
      "eval_steps_per_second": 0.879,
      "step": 400
    },
    {
      "epoch": 50.0,
      "learning_rate": 5e-06,
      "loss": 0.2429,
      "step": 500
    },
    {
      "epoch": 50.0,
      "eval_accuracy": 0.8298116674322462,
      "eval_f1": 0.9014154457412467,
      "eval_loss": 1.001568078994751,
      "eval_precision": 0.895411938825851,
      "eval_recall": 0.9075,
      "eval_runtime": 4.6434,
      "eval_samples_per_second": 11.629,
      "eval_steps_per_second": 0.861,
      "step": 500
    },
    {
      "epoch": 60.0,
      "eval_accuracy": 0.8293523197060174,
      "eval_f1": 0.8934262948207171,
      "eval_loss": 1.009719967842102,
      "eval_precision": 0.8898809523809523,
      "eval_recall": 0.897,
      "eval_runtime": 4.6541,
      "eval_samples_per_second": 11.603,
      "eval_steps_per_second": 0.859,
      "step": 600
    },
    {
      "epoch": 70.0,
      "eval_accuracy": 0.8315342214056041,
      "eval_f1": 0.906008476689105,
      "eval_loss": 1.072226643562317,
      "eval_precision": 0.9035305818000995,
      "eval_recall": 0.9085,
      "eval_runtime": 4.5145,
      "eval_samples_per_second": 11.961,
      "eval_steps_per_second": 0.886,
      "step": 700
    },
    {
      "epoch": 80.0,
      "eval_accuracy": 0.8269407441433165,
      "eval_f1": 0.9003708281829418,
      "eval_loss": 1.088362693786621,
      "eval_precision": 0.8904645476772616,
      "eval_recall": 0.9105,
      "eval_runtime": 4.4141,
      "eval_samples_per_second": 12.233,
      "eval_steps_per_second": 0.906,
      "step": 800
    },
    {
      "epoch": 90.0,
      "eval_accuracy": 0.827859439595774,
      "eval_f1": 0.9013386217154189,
      "eval_loss": 1.129175066947937,
      "eval_precision": 0.8938053097345132,
      "eval_recall": 0.909,
      "eval_runtime": 4.4449,
      "eval_samples_per_second": 12.149,
      "eval_steps_per_second": 0.9,
      "step": 900
    },
    {
      "epoch": 100.0,
      "learning_rate": 0.0,
      "loss": 0.0098,
      "step": 1000
    },
    {
      "epoch": 100.0,
      "eval_accuracy": 0.8330271015158475,
      "eval_f1": 0.9077802634849614,
      "eval_loss": 1.1164220571517944,
      "eval_precision": 0.9026198714780029,
      "eval_recall": 0.913,
      "eval_runtime": 4.533,
      "eval_samples_per_second": 11.913,
      "eval_steps_per_second": 0.882,
      "step": 1000
    },
    {
      "epoch": 100.0,
      "step": 1000,
      "total_flos": 3954080148480000.0,
      "train_loss": 0.12633917331695557,
      "train_runtime": 1653.606,
      "train_samples_per_second": 9.676,
      "train_steps_per_second": 0.605
    }
  ],
  "max_steps": 1000,
  "num_train_epochs": 100,
  "total_flos": 3954080148480000.0,
  "trial_name": null,
  "trial_params": null
}