trainer_state.json

{
  "best_metric": 0.8781333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0819/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": 4.376876876876877e-05,
      "loss": 2.093,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7530666666666667,
      "eval_loss": 0.9361585378646851,
      "eval_runtime": 4.6368,
      "eval_samples_per_second": 808.741,
      "eval_steps_per_second": 12.724,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 3.7537537537537536e-05,
      "loss": 0.6386,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8264,
      "eval_loss": 0.6119039058685303,
      "eval_runtime": 4.1102,
      "eval_samples_per_second": 912.372,
      "eval_steps_per_second": 14.355,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.41261613368988037,
      "learning_rate": 3.128753753753754e-05,
      "loss": 0.3508,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8549333333333333,
      "eval_loss": 0.49816203117370605,
      "eval_runtime": 4.6648,
      "eval_samples_per_second": 803.889,
      "eval_steps_per_second": 12.648,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 6.447729110717773,
      "learning_rate": 2.5037537537537537e-05,
      "loss": 0.2073,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8610666666666666,
      "eval_loss": 0.4731382131576538,
      "eval_runtime": 5.4852,
      "eval_samples_per_second": 683.663,
      "eval_steps_per_second": 10.756,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.13128192722797394,
      "learning_rate": 1.8787537537537537e-05,
      "loss": 0.1191,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8634666666666667,
      "eval_loss": 0.4631704092025757,
      "eval_runtime": 4.564,
      "eval_samples_per_second": 821.65,
      "eval_steps_per_second": 12.927,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 7.007894515991211,
      "learning_rate": 1.2537537537537538e-05,
      "loss": 0.0658,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8661333333333333,
      "eval_loss": 0.4570779800415039,
      "eval_runtime": 4.1915,
      "eval_samples_per_second": 894.664,
      "eval_steps_per_second": 14.076,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.20807838439941406,
      "learning_rate": 6.287537537537539e-06,
      "loss": 0.0345,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8746666666666667,
      "eval_loss": 0.4598044157028198,
      "eval_runtime": 4.0594,
      "eval_samples_per_second": 923.775,
      "eval_steps_per_second": 14.534,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.5788722038269043,
      "learning_rate": 3.753753753753754e-08,
      "loss": 0.0201,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8781333333333333,
      "eval_loss": 0.4508090615272522,
      "eval_runtime": 4.0808,
      "eval_samples_per_second": 918.934,
      "eval_steps_per_second": 14.458,
      "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
}