trainer_state.json

{
  "best_metric": 0.9301333333333334,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0386/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": 4.811948852959657e-05,
      "loss": 0.7569,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8768,
      "eval_loss": 0.4110783338546753,
      "eval_runtime": 7.0774,
      "eval_samples_per_second": 529.857,
      "eval_steps_per_second": 8.336,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 4.274009935225696e-05,
      "loss": 0.2435,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8861333333333333,
      "eval_loss": 0.3836166560649872,
      "eval_runtime": 7.0325,
      "eval_samples_per_second": 533.239,
      "eval_steps_per_second": 8.39,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 5.063878059387207,
      "learning_rate": 3.464873897488322e-05,
      "loss": 0.106,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8989333333333334,
      "eval_loss": 0.3477427661418915,
      "eval_runtime": 8.104,
      "eval_samples_per_second": 462.733,
      "eval_steps_per_second": 7.28,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.006217686925083399,
      "learning_rate": 2.5088445554620133e-05,
      "loss": 0.055,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.896,
      "eval_loss": 0.3758491575717926,
      "eval_runtime": 6.4742,
      "eval_samples_per_second": 579.222,
      "eval_steps_per_second": 9.113,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 10.947654724121094,
      "learning_rate": 1.5514687100427083e-05,
      "loss": 0.0248,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9128,
      "eval_loss": 0.344719797372818,
      "eval_runtime": 6.3429,
      "eval_samples_per_second": 591.208,
      "eval_steps_per_second": 9.302,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.025384292006492615,
      "learning_rate": 7.384981550618447e-06,
      "loss": 0.0079,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9288,
      "eval_loss": 0.31276023387908936,
      "eval_runtime": 6.512,
      "eval_samples_per_second": 575.864,
      "eval_steps_per_second": 9.06,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.0012686976697295904,
      "learning_rate": 1.937002879188285e-06,
      "loss": 0.0022,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.928,
      "eval_loss": 0.3169269263744354,
      "eval_runtime": 7.0137,
      "eval_samples_per_second": 534.669,
      "eval_steps_per_second": 8.412,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.0520748607814312,
      "learning_rate": 1.564528121916764e-10,
      "loss": 0.0014,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9301333333333334,
      "eval_loss": 0.30875009298324585,
      "eval_runtime": 6.7281,
      "eval_samples_per_second": 557.365,
      "eval_steps_per_second": 8.769,
      "step": 2664
    }
  ],
  "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": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.317930564980736e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}