trainer_state.json

{
  "best_metric": 0.9296,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0138/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": 0.0001,
      "loss": 0.6428,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9258666666666666,
      "eval_loss": 0.24224084615707397,
      "eval_runtime": 8.3965,
      "eval_samples_per_second": 446.617,
      "eval_steps_per_second": 7.027,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.19827118515968323,
      "learning_rate": 0.0001,
      "loss": 0.1669,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9274666666666667,
      "eval_loss": 0.238366961479187,
      "eval_runtime": 7.4775,
      "eval_samples_per_second": 501.505,
      "eval_steps_per_second": 7.89,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 47.68202590942383,
      "learning_rate": 0.0001,
      "loss": 0.0938,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9229333333333334,
      "eval_loss": 0.27570730447769165,
      "eval_runtime": 7.4583,
      "eval_samples_per_second": 502.795,
      "eval_steps_per_second": 7.911,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 2.3839898109436035,
      "learning_rate": 0.0001,
      "loss": 0.0663,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9058666666666667,
      "eval_loss": 0.3448584973812103,
      "eval_runtime": 7.485,
      "eval_samples_per_second": 501.003,
      "eval_steps_per_second": 7.882,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.027057982981204987,
      "learning_rate": 0.0001,
      "loss": 0.0568,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9173333333333333,
      "eval_loss": 0.3414529860019684,
      "eval_runtime": 7.072,
      "eval_samples_per_second": 530.26,
      "eval_steps_per_second": 8.343,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.020032014697790146,
      "learning_rate": 0.0001,
      "loss": 0.0422,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9218666666666666,
      "eval_loss": 0.323521226644516,
      "eval_runtime": 7.414,
      "eval_samples_per_second": 505.799,
      "eval_steps_per_second": 7.958,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.008711430244147778,
      "learning_rate": 0.0001,
      "loss": 0.0409,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9218666666666666,
      "eval_loss": 0.3256213665008545,
      "eval_runtime": 10.113,
      "eval_samples_per_second": 370.809,
      "eval_steps_per_second": 5.834,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.009592837654054165,
      "learning_rate": 0.0001,
      "loss": 0.0399,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9296,
      "eval_loss": 0.29305005073547363,
      "eval_runtime": 7.4307,
      "eval_samples_per_second": 504.662,
      "eval_steps_per_second": 7.94,
      "step": 2664
    }
  ],
  "logging_steps": 500,
  "max_steps": 2997,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9,
  "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": 1.317930564980736e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}