trainer_state.json

{
  "best_metric": 0.8392,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0776/checkpoints/checkpoint-2664",
  "epoch": 8.0,
  "eval_steps": 500,
  "global_step": 2664,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": Infinity,
      "learning_rate": 0.0002910075938432782,
      "loss": 2.4591,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5986666666666667,
      "eval_loss": 1.3239847421646118,
      "eval_runtime": 6.6959,
      "eval_samples_per_second": 560.042,
      "eval_steps_per_second": 8.811,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 4.882388114929199,
      "learning_rate": 0.0002650076732300463,
      "loss": 0.9389,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7288,
      "eval_loss": 0.8787519335746765,
      "eval_runtime": 5.7279,
      "eval_samples_per_second": 654.696,
      "eval_steps_per_second": 10.301,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.7930739521980286,
      "learning_rate": 0.00022513612989278497,
      "loss": 0.4786,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7474666666666666,
      "eval_loss": 0.831194281578064,
      "eval_runtime": 6.1536,
      "eval_samples_per_second": 609.397,
      "eval_steps_per_second": 9.588,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.5209512114524841,
      "learning_rate": 0.00017620206039927676,
      "loss": 0.2201,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7858666666666667,
      "eval_loss": 0.7246662974357605,
      "eval_runtime": 6.4514,
      "eval_samples_per_second": 581.27,
      "eval_steps_per_second": 9.145,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.07298845052719116,
      "learning_rate": 0.00012410763572038907,
      "loss": 0.0875,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.7984,
      "eval_loss": 0.7305771112442017,
      "eval_runtime": 5.665,
      "eval_samples_per_second": 661.965,
      "eval_steps_per_second": 10.415,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.8890611529350281,
      "learning_rate": 7.513621230422122e-05,
      "loss": 0.025,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8149333333333333,
      "eval_loss": 0.6994869709014893,
      "eval_runtime": 5.738,
      "eval_samples_per_second": 653.536,
      "eval_steps_per_second": 10.282,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.013389199040830135,
      "learning_rate": 3.519446655599853e-05,
      "loss": 0.0062,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8354666666666667,
      "eval_loss": 0.6733562350273132,
      "eval_runtime": 5.6891,
      "eval_samples_per_second": 659.151,
      "eval_steps_per_second": 10.371,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.011826746165752411,
      "learning_rate": 9.099962490342783e-06,
      "loss": 0.0025,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8392,
      "eval_loss": 0.6628664135932922,
      "eval_runtime": 5.6302,
      "eval_samples_per_second": 666.05,
      "eval_steps_per_second": 10.479,
      "step": 2664
    }
  ],
  "logging_steps": 500,
  "max_steps": 2997,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9,
  "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": 1.317930564980736e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}