trainer_state.json

{
  "best_metric": 0.5357333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0016/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": 0.0004850126564054637,
      "loss": 3.3128,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.21226666666666666,
      "eval_loss": 2.90122389793396,
      "eval_runtime": 6.5353,
      "eval_samples_per_second": 573.804,
      "eval_steps_per_second": 9.028,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 0.00044184758936602033,
      "loss": 2.7332,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.31093333333333334,
      "eval_loss": 2.523120403289795,
      "eval_runtime": 5.8701,
      "eval_samples_per_second": 638.832,
      "eval_steps_per_second": 10.051,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": Infinity,
      "learning_rate": 0.00037568023640968044,
      "loss": 2.3539,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.39386666666666664,
      "eval_loss": 2.191303014755249,
      "eval_runtime": 5.6266,
      "eval_samples_per_second": 666.479,
      "eval_steps_per_second": 10.486,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 6.701728343963623,
      "learning_rate": 0.00029418606892278544,
      "loss": 1.9976,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.4362666666666667,
      "eval_loss": 2.046833038330078,
      "eval_runtime": 5.6761,
      "eval_samples_per_second": 660.663,
      "eval_steps_per_second": 10.394,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 13.051854133605957,
      "learning_rate": 0.0002073624094068776,
      "loss": 1.6545,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.49546666666666667,
      "eval_loss": 1.8192414045333862,
      "eval_runtime": 5.672,
      "eval_samples_per_second": 661.148,
      "eval_steps_per_second": 10.402,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 22.20870018005371,
      "learning_rate": 0.00012568147258031897,
      "loss": 1.2646,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.5184,
      "eval_loss": 1.7464109659194946,
      "eval_runtime": 5.6557,
      "eval_samples_per_second": 663.048,
      "eval_steps_per_second": 10.432,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 2.7283196449279785,
      "learning_rate": 5.899518490663266e-05,
      "loss": 0.8065,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.5333333333333333,
      "eval_loss": 1.7455294132232666,
      "eval_runtime": 5.6924,
      "eval_samples_per_second": 658.775,
      "eval_steps_per_second": 10.365,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 5.679820537567139,
      "learning_rate": 1.534689686405272e-05,
      "loss": 0.4057,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.5357333333333333,
      "eval_loss": 1.7825565338134766,
      "eval_runtime": 5.7274,
      "eval_samples_per_second": 654.751,
      "eval_steps_per_second": 10.301,
      "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
}