trainer_state.json

{
  "best_metric": 0.06341079622507095,
  "best_model_checkpoint": "vit-large-patch16-224-new-dungeon-geo-morphs-22Nov24-005/checkpoint-30",
  "epoch": 20.0,
  "eval_steps": 10,
  "global_step": 30,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 6.666666666666667,
      "grad_norm": 7.522892475128174,
      "learning_rate": 1.555555555555556e-05,
      "loss": 0.8777,
      "step": 10
    },
    {
      "epoch": 6.666666666666667,
      "eval_accuracy": 0.9259259259259259,
      "eval_loss": 0.3644285202026367,
      "eval_runtime": 0.6294,
      "eval_samples_per_second": 42.896,
      "eval_steps_per_second": 6.355,
      "step": 10
    },
    {
      "epoch": 13.333333333333334,
      "grad_norm": 0.2936255931854248,
      "learning_rate": 8.148148148148148e-06,
      "loss": 0.0406,
      "step": 20
    },
    {
      "epoch": 13.333333333333334,
      "eval_accuracy": 1.0,
      "eval_loss": 0.09418925642967224,
      "eval_runtime": 0.4619,
      "eval_samples_per_second": 58.451,
      "eval_steps_per_second": 8.659,
      "step": 20
    },
    {
      "epoch": 20.0,
      "grad_norm": 0.08716100454330444,
      "learning_rate": 7.407407407407407e-07,
      "loss": 0.0021,
      "step": 30
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 1.0,
      "eval_loss": 0.06341079622507095,
      "eval_runtime": 0.5383,
      "eval_samples_per_second": 50.158,
      "eval_steps_per_second": 7.431,
      "step": 30
    },
    {
      "epoch": 20.0,
      "step": 30,
      "total_flos": 2.6297859010461696e+17,
      "train_loss": 0.3068043768405914,
      "train_runtime": 142.4913,
      "train_samples_per_second": 10.106,
      "train_steps_per_second": 0.211
    }
  ],
  "logging_steps": 10,
  "max_steps": 30,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 30,
  "save_steps": 10,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 2.6297859010461696e+17,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}