trainer_state.json

{
  "best_metric": 0.9,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0515/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": Infinity,
      "learning_rate": 8.659485808097889e-05,
      "loss": 1.7117,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8026666666666666,
      "eval_loss": 0.7061209678649902,
      "eval_runtime": 10.412,
      "eval_samples_per_second": 360.163,
      "eval_steps_per_second": 5.667,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 27.990272521972656,
      "learning_rate": 7.685730736804761e-05,
      "loss": 0.5159,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8322666666666667,
      "eval_loss": 0.5694859623908997,
      "eval_runtime": 6.8479,
      "eval_samples_per_second": 547.612,
      "eval_steps_per_second": 8.616,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.615123987197876,
      "learning_rate": 6.226977039554153e-05,
      "loss": 0.2857,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8552,
      "eval_loss": 0.493453711271286,
      "eval_runtime": 6.8194,
      "eval_samples_per_second": 549.905,
      "eval_steps_per_second": 8.652,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 3.6490139961242676,
      "learning_rate": 4.505306743117272e-05,
      "loss": 0.1569,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8696,
      "eval_loss": 0.45199277997016907,
      "eval_runtime": 7.2278,
      "eval_samples_per_second": 518.83,
      "eval_steps_per_second": 8.163,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.6708877682685852,
      "learning_rate": 2.7828285431465317e-05,
      "loss": 0.0774,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8818666666666667,
      "eval_loss": 0.42081764340400696,
      "eval_runtime": 6.7843,
      "eval_samples_per_second": 552.749,
      "eval_steps_per_second": 8.697,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.05931715667247772,
      "learning_rate": 1.3217741312837138e-05,
      "loss": 0.034,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8949333333333334,
      "eval_loss": 0.4008956849575043,
      "eval_runtime": 6.8962,
      "eval_samples_per_second": 543.781,
      "eval_steps_per_second": 8.555,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.05061155557632446,
      "learning_rate": 3.4457579724370674e-06,
      "loss": 0.0148,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9,
      "eval_loss": 0.378093957901001,
      "eval_runtime": 6.7293,
      "eval_samples_per_second": 557.261,
      "eval_steps_per_second": 8.768,
      "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
}