trainer_state.json

{
  "best_metric": 0.14869779348373413,
  "best_model_checkpoint": "vit-large-patch16-224-dungeon-geo-morphs-012/checkpoint-30",
  "epoch": 23.285714285714285,
  "eval_steps": 10,
  "global_step": 35,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 6.571428571428571,
      "grad_norm": 0.4156835973262787,
      "learning_rate": 1.6129032258064517e-05,
      "loss": 0.0,
      "step": 10
    },
    {
      "epoch": 6.571428571428571,
      "eval_accuracy": 0.9722222222222222,
      "eval_loss": 0.16624921560287476,
      "eval_runtime": 0.5515,
      "eval_samples_per_second": 65.271,
      "eval_steps_per_second": 9.065,
      "step": 10
    },
    {
      "epoch": 13.285714285714286,
      "grad_norm": 0.001085005234926939,
      "learning_rate": 1.0322580645161291e-05,
      "loss": 0.0044,
      "step": 20
    },
    {
      "epoch": 13.285714285714286,
      "eval_accuracy": 0.9444444444444444,
      "eval_loss": 0.22176334261894226,
      "eval_runtime": 0.5681,
      "eval_samples_per_second": 63.37,
      "eval_steps_per_second": 8.801,
      "step": 20
    },
    {
      "epoch": 19.857142857142858,
      "grad_norm": 0.000658858974929899,
      "learning_rate": 3.870967741935484e-06,
      "loss": 0.0,
      "step": 30
    },
    {
      "epoch": 19.857142857142858,
      "eval_accuracy": 0.9722222222222222,
      "eval_loss": 0.14869779348373413,
      "eval_runtime": 0.5918,
      "eval_samples_per_second": 60.834,
      "eval_steps_per_second": 8.449,
      "step": 30
    },
    {
      "epoch": 23.285714285714285,
      "step": 35,
      "total_flos": 3.068073182851891e+17,
      "train_loss": 0.0012724436819553375,
      "train_runtime": 193.1443,
      "train_samples_per_second": 9.061,
      "train_steps_per_second": 0.181
    }
  ],
  "logging_steps": 10,
  "max_steps": 35,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 35,
  "save_steps": 10,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 5,
        "early_stopping_threshold": 0.01
      },
      "attributes": {
        "early_stopping_patience_counter": 0
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 3.068073182851891e+17,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}