trainer_state.json

{
  "best_metric": 0.5405333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0886/checkpoints/checkpoint-1665",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 1665,
  "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": 0.0005,
      "loss": 3.0009,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.2770666666666667,
      "eval_loss": 2.626103639602661,
      "eval_runtime": 6.7979,
      "eval_samples_per_second": 551.637,
      "eval_steps_per_second": 8.679,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 0.0005,
      "loss": 2.2574,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.4104,
      "eval_loss": 2.093247413635254,
      "eval_runtime": 6.2286,
      "eval_samples_per_second": 602.066,
      "eval_steps_per_second": 9.473,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 12.349518775939941,
      "learning_rate": 0.0005,
      "loss": 1.8261,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.47786666666666666,
      "eval_loss": 1.816083312034607,
      "eval_runtime": 6.7553,
      "eval_samples_per_second": 555.119,
      "eval_steps_per_second": 8.734,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 14.414395332336426,
      "learning_rate": 0.0005,
      "loss": 1.4859,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.5370666666666667,
      "eval_loss": 1.6274113655090332,
      "eval_runtime": 6.2125,
      "eval_samples_per_second": 603.623,
      "eval_steps_per_second": 9.497,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 10.039994239807129,
      "learning_rate": 0.0005,
      "loss": 1.1809,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.5405333333333333,
      "eval_loss": 1.594435453414917,
      "eval_runtime": 6.2849,
      "eval_samples_per_second": 596.671,
      "eval_steps_per_second": 9.388,
      "step": 1665
    }
  ],
  "logging_steps": 500,
  "max_steps": 1665,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "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": 8.2370660311296e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}