trainer_state.json

{
  "best_metric": 0.4461333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0562/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": 0.0003,
      "loss": 3.203,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.24506666666666665,
      "eval_loss": 2.7688894271850586,
      "eval_runtime": 11.9023,
      "eval_samples_per_second": 315.064,
      "eval_steps_per_second": 4.957,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 0.0003,
      "loss": 2.638,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.3152,
      "eval_loss": 2.487724542617798,
      "eval_runtime": 11.7373,
      "eval_samples_per_second": 319.494,
      "eval_steps_per_second": 5.027,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 15.478167533874512,
      "learning_rate": 0.0003,
      "loss": 2.3916,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.3376,
      "eval_loss": 2.390247344970703,
      "eval_runtime": 11.3741,
      "eval_samples_per_second": 329.698,
      "eval_steps_per_second": 5.187,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 20.72833251953125,
      "learning_rate": 0.0003,
      "loss": 2.225,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.3664,
      "eval_loss": 2.2711710929870605,
      "eval_runtime": 11.6544,
      "eval_samples_per_second": 321.767,
      "eval_steps_per_second": 5.062,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 20.960500717163086,
      "learning_rate": 0.0003,
      "loss": 2.082,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.3944,
      "eval_loss": 2.207716941833496,
      "eval_runtime": 11.5968,
      "eval_samples_per_second": 323.366,
      "eval_steps_per_second": 5.088,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 26.33807373046875,
      "learning_rate": 0.0003,
      "loss": 1.9617,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.4338666666666667,
      "eval_loss": 2.037109613418579,
      "eval_runtime": 11.5325,
      "eval_samples_per_second": 325.169,
      "eval_steps_per_second": 5.116,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 12.124343872070312,
      "learning_rate": 0.0003,
      "loss": 1.8405,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.4384,
      "eval_loss": 2.0238709449768066,
      "eval_runtime": 11.5872,
      "eval_samples_per_second": 323.632,
      "eval_steps_per_second": 5.092,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 25.658376693725586,
      "learning_rate": 0.0003,
      "loss": 1.7279,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.4461333333333333,
      "eval_loss": 1.9898390769958496,
      "eval_runtime": 11.5281,
      "eval_samples_per_second": 325.293,
      "eval_steps_per_second": 5.118,
      "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
}