trainer_state.json

{
  "best_metric": 0.8717333333333334,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0763/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": 5e-05,
      "loss": 1.7438,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7866666666666666,
      "eval_loss": 0.8156900405883789,
      "eval_runtime": 6.4009,
      "eval_samples_per_second": 585.852,
      "eval_steps_per_second": 9.217,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 5e-05,
      "loss": 0.5106,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.848,
      "eval_loss": 0.5349177122116089,
      "eval_runtime": 6.2992,
      "eval_samples_per_second": 595.312,
      "eval_steps_per_second": 9.366,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 4.058676242828369,
      "learning_rate": 5e-05,
      "loss": 0.2673,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8586666666666667,
      "eval_loss": 0.48665598034858704,
      "eval_runtime": 6.8885,
      "eval_samples_per_second": 544.385,
      "eval_steps_per_second": 8.565,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 4.887714862823486,
      "learning_rate": 5e-05,
      "loss": 0.157,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8701333333333333,
      "eval_loss": 0.4854186773300171,
      "eval_runtime": 6.3553,
      "eval_samples_per_second": 590.059,
      "eval_steps_per_second": 9.284,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 3.219672918319702,
      "learning_rate": 5e-05,
      "loss": 0.0996,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8669333333333333,
      "eval_loss": 0.4746582806110382,
      "eval_runtime": 6.2905,
      "eval_samples_per_second": 596.141,
      "eval_steps_per_second": 9.379,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.1400194764137268,
      "learning_rate": 5e-05,
      "loss": 0.0699,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8677333333333334,
      "eval_loss": 0.5177216529846191,
      "eval_runtime": 6.4325,
      "eval_samples_per_second": 582.978,
      "eval_steps_per_second": 9.172,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 5.555139064788818,
      "learning_rate": 5e-05,
      "loss": 0.0559,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8717333333333334,
      "eval_loss": 0.5231451392173767,
      "eval_runtime": 7.7792,
      "eval_samples_per_second": 482.057,
      "eval_steps_per_second": 7.584,
      "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
}