trainer_state.json

{
  "best_metric": 0.8506666666666667,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0548/checkpoints/checkpoint-1998",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 1998,
  "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": 0.0001,
      "loss": 1.8633,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7728,
      "eval_loss": 0.8075184226036072,
      "eval_runtime": 5.586,
      "eval_samples_per_second": 671.322,
      "eval_steps_per_second": 10.562,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 4.956782341003418,
      "learning_rate": 0.0001,
      "loss": 0.5334,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8165333333333333,
      "eval_loss": 0.5839399695396423,
      "eval_runtime": 5.6003,
      "eval_samples_per_second": 669.612,
      "eval_steps_per_second": 10.535,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.3710747957229614,
      "learning_rate": 0.0001,
      "loss": 0.2895,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8501333333333333,
      "eval_loss": 0.5089996457099915,
      "eval_runtime": 5.5128,
      "eval_samples_per_second": 680.24,
      "eval_steps_per_second": 10.702,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": Infinity,
      "learning_rate": 0.0001,
      "loss": 0.1769,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8376,
      "eval_loss": 0.5848579406738281,
      "eval_runtime": 5.7922,
      "eval_samples_per_second": 647.419,
      "eval_steps_per_second": 10.186,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 5.304078578948975,
      "learning_rate": 0.0001,
      "loss": 0.1222,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.848,
      "eval_loss": 0.5741931200027466,
      "eval_runtime": 5.5766,
      "eval_samples_per_second": 672.457,
      "eval_steps_per_second": 10.58,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 10.515223503112793,
      "learning_rate": 0.0001,
      "loss": 0.086,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8506666666666667,
      "eval_loss": 0.5734532475471497,
      "eval_runtime": 5.6802,
      "eval_samples_per_second": 660.188,
      "eval_steps_per_second": 10.387,
      "step": 1998
    }
  ],
  "logging_steps": 500,
  "max_steps": 2331,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "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": 9.88447923735552e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}