trainer_state.json

{
  "best_metric": 0.9272,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0940/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": 6.226893560226894e-05,
      "loss": 0.7751,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8549333333333333,
      "eval_loss": 0.47476157546043396,
      "eval_runtime": 11.1887,
      "eval_samples_per_second": 335.16,
      "eval_steps_per_second": 5.273,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 80.75167083740234,
      "learning_rate": 5.449115782449115e-05,
      "loss": 0.2347,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8658666666666667,
      "eval_loss": 0.43757036328315735,
      "eval_runtime": 10.7966,
      "eval_samples_per_second": 347.332,
      "eval_steps_per_second": 5.465,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 5.857204914093018,
      "learning_rate": 4.671338004671337e-05,
      "loss": 0.1129,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8944,
      "eval_loss": 0.38149821758270264,
      "eval_runtime": 11.0036,
      "eval_samples_per_second": 340.797,
      "eval_steps_per_second": 5.362,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 18.306970596313477,
      "learning_rate": 3.89356022689356e-05,
      "loss": 0.0539,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9013333333333333,
      "eval_loss": 0.3655620217323303,
      "eval_runtime": 10.8555,
      "eval_samples_per_second": 345.448,
      "eval_steps_per_second": 5.435,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.15495896339416504,
      "learning_rate": 3.115782449115782e-05,
      "loss": 0.0335,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9058666666666667,
      "eval_loss": 0.3944179117679596,
      "eval_runtime": 10.8353,
      "eval_samples_per_second": 346.092,
      "eval_steps_per_second": 5.445,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.0007641515112482011,
      "learning_rate": 2.3380046713380048e-05,
      "loss": 0.0127,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.916,
      "eval_loss": 0.36178314685821533,
      "eval_runtime": 10.804,
      "eval_samples_per_second": 347.094,
      "eval_steps_per_second": 5.461,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.005974370986223221,
      "learning_rate": 1.5602268935602267e-05,
      "loss": 0.0054,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9218666666666666,
      "eval_loss": 0.348127156496048,
      "eval_runtime": 10.7793,
      "eval_samples_per_second": 347.889,
      "eval_steps_per_second": 5.473,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.0003166375681757927,
      "learning_rate": 7.824491157824491e-06,
      "loss": 0.0024,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9272,
      "eval_loss": 0.33199256658554077,
      "eval_runtime": 10.8289,
      "eval_samples_per_second": 346.295,
      "eval_steps_per_second": 5.448,
      "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
}