trainer_state.json

{
  "best_metric": 0.9637654090399701,
  "best_model_checkpoint": "layoutlmv3-test-cord/checkpoint-1000",
  "epoch": 20.0,
  "global_step": 1000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8709677419354839,
      "eval_f1": 0.838591342626559,
      "eval_loss": 0.5257240533828735,
      "eval_precision": 0.8223021582733813,
      "eval_recall": 0.8555389221556886,
      "eval_runtime": 8.1118,
      "eval_samples_per_second": 12.328,
      "eval_steps_per_second": 0.863,
      "step": 100
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9316638370118846,
      "eval_f1": 0.9198813056379822,
      "eval_loss": 0.3200061023235321,
      "eval_precision": 0.9117647058823529,
      "eval_recall": 0.9281437125748503,
      "eval_runtime": 7.9991,
      "eval_samples_per_second": 12.501,
      "eval_steps_per_second": 0.875,
      "step": 200
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9465195246179966,
      "eval_f1": 0.9360594795539035,
      "eval_loss": 0.2448921650648117,
      "eval_precision": 0.9298375184638109,
      "eval_recall": 0.9423652694610778,
      "eval_runtime": 7.8919,
      "eval_samples_per_second": 12.671,
      "eval_steps_per_second": 0.887,
      "step": 300
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9596774193548387,
      "eval_f1": 0.9503916449086163,
      "eval_loss": 0.19232001900672913,
      "eval_precision": 0.9472118959107807,
      "eval_recall": 0.9535928143712575,
      "eval_runtime": 8.0551,
      "eval_samples_per_second": 12.415,
      "eval_steps_per_second": 0.869,
      "step": 400
    },
    {
      "epoch": 10.0,
      "learning_rate": 2.5e-05,
      "loss": 0.4328,
      "step": 500
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.9681663837011885,
      "eval_f1": 0.9623274897426333,
      "eval_loss": 0.18568018078804016,
      "eval_precision": 0.9591078066914498,
      "eval_recall": 0.9655688622754491,
      "eval_runtime": 8.1741,
      "eval_samples_per_second": 12.234,
      "eval_steps_per_second": 0.856,
      "step": 500
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.9656196943972836,
      "eval_f1": 0.9607476635514018,
      "eval_loss": 0.20726947486400604,
      "eval_precision": 0.9596713965646004,
      "eval_recall": 0.9618263473053892,
      "eval_runtime": 8.2339,
      "eval_samples_per_second": 12.145,
      "eval_steps_per_second": 0.85,
      "step": 600
    },
    {
      "epoch": 14.0,
      "eval_accuracy": 0.9702886247877759,
      "eval_f1": 0.9648729446935724,
      "eval_loss": 0.1803944706916809,
      "eval_precision": 0.9634328358208956,
      "eval_recall": 0.9663173652694611,
      "eval_runtime": 7.9827,
      "eval_samples_per_second": 12.527,
      "eval_steps_per_second": 0.877,
      "step": 700
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.9664685908319185,
      "eval_f1": 0.9640987284966342,
      "eval_loss": 0.18818138539791107,
      "eval_precision": 0.9633781763826607,
      "eval_recall": 0.9648203592814372,
      "eval_runtime": 8.0654,
      "eval_samples_per_second": 12.399,
      "eval_steps_per_second": 0.868,
      "step": 800
    },
    {
      "epoch": 18.0,
      "eval_accuracy": 0.967741935483871,
      "eval_f1": 0.9633781763826607,
      "eval_loss": 0.1799575388431549,
      "eval_precision": 0.9619402985074627,
      "eval_recall": 0.9648203592814372,
      "eval_runtime": 7.9449,
      "eval_samples_per_second": 12.587,
      "eval_steps_per_second": 0.881,
      "step": 900
    },
    {
      "epoch": 20.0,
      "learning_rate": 0.0,
      "loss": 0.0318,
      "step": 1000
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 0.9681663837011885,
      "eval_f1": 0.9637654090399701,
      "eval_loss": 0.18448814749717712,
      "eval_precision": 0.9619686800894854,
      "eval_recall": 0.9655688622754491,
      "eval_runtime": 8.0126,
      "eval_samples_per_second": 12.48,
      "eval_steps_per_second": 0.874,
      "step": 1000
    },
    {
      "epoch": 20.0,
      "step": 1000,
      "total_flos": 4248755355648000.0,
      "train_loss": 0.2323043212890625,
      "train_runtime": 1807.2991,
      "train_samples_per_second": 8.853,
      "train_steps_per_second": 0.553
    }
  ],
  "max_steps": 1000,
  "num_train_epochs": 20,
  "total_flos": 4248755355648000.0,
  "trial_name": null,
  "trial_params": null
}