trainer_state.json

{
  "best_metric": 0.8701333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0645/checkpoints/checkpoint-2331",
  "epoch": 7.0,
  "eval_steps": 500,
  "global_step": 2331,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 3.1647114753723145,
      "learning_rate": 5e-05,
      "loss": 1.8811,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7917333333333333,
      "eval_loss": 0.8049337267875671,
      "eval_runtime": 7.5036,
      "eval_samples_per_second": 499.763,
      "eval_steps_per_second": 7.863,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 5e-05,
      "loss": 0.5859,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8426666666666667,
      "eval_loss": 0.5559996366500854,
      "eval_runtime": 8.4129,
      "eval_samples_per_second": 445.744,
      "eval_steps_per_second": 7.013,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 28.457807540893555,
      "learning_rate": 5e-05,
      "loss": 0.3453,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8552,
      "eval_loss": 0.4958791732788086,
      "eval_runtime": 7.4606,
      "eval_samples_per_second": 502.638,
      "eval_steps_per_second": 7.908,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.34068575501441956,
      "learning_rate": 5e-05,
      "loss": 0.2311,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8530666666666666,
      "eval_loss": 0.5079152584075928,
      "eval_runtime": 7.6128,
      "eval_samples_per_second": 492.589,
      "eval_steps_per_second": 7.75,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 3.240177631378174,
      "learning_rate": 5e-05,
      "loss": 0.1651,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8634666666666667,
      "eval_loss": 0.48043835163116455,
      "eval_runtime": 8.7815,
      "eval_samples_per_second": 427.032,
      "eval_steps_per_second": 6.719,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.25386080145835876,
      "learning_rate": 5e-05,
      "loss": 0.1223,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8672,
      "eval_loss": 0.4800771474838257,
      "eval_runtime": 7.7126,
      "eval_samples_per_second": 486.218,
      "eval_steps_per_second": 7.65,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.8602979779243469,
      "learning_rate": 5e-05,
      "loss": 0.0896,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8701333333333333,
      "eval_loss": 0.4872928857803345,
      "eval_runtime": 7.3462,
      "eval_samples_per_second": 510.466,
      "eval_steps_per_second": 8.031,
      "step": 2331
    }
  ],
  "logging_steps": 500,
  "max_steps": 2664,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.153189244358144e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}