trainer_state.json

{
  "best_metric": 0.9085333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0375/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": 9.14895248413086,
      "learning_rate": 0.0004809698831278217,
      "loss": 0.808,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.768,
      "eval_loss": 0.8071669340133667,
      "eval_runtime": 7.1945,
      "eval_samples_per_second": 521.232,
      "eval_steps_per_second": 8.201,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 0.0004269850409335979,
      "loss": 0.464,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8253333333333334,
      "eval_loss": 0.6033662557601929,
      "eval_runtime": 7.2858,
      "eval_samples_per_second": 514.702,
      "eval_steps_per_second": 8.098,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 8.1299409866333,
      "learning_rate": 0.0003459431688641196,
      "loss": 0.3007,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8285333333333333,
      "eval_loss": 0.5847275853157043,
      "eval_runtime": 7.1372,
      "eval_samples_per_second": 525.415,
      "eval_steps_per_second": 8.267,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.0652560219168663,
      "learning_rate": 0.00025029481906207064,
      "loss": 0.1879,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8546666666666667,
      "eval_loss": 0.5422940254211426,
      "eval_runtime": 7.1979,
      "eval_samples_per_second": 520.988,
      "eval_steps_per_second": 8.197,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 3.3689959049224854,
      "learning_rate": 0.00015460158573036287,
      "loss": 0.0932,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8792,
      "eval_loss": 0.44306617975234985,
      "eval_runtime": 7.4822,
      "eval_samples_per_second": 501.192,
      "eval_steps_per_second": 7.885,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 3.3317956924438477,
      "learning_rate": 7.343189618242854e-05,
      "loss": 0.0357,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8906666666666667,
      "eval_loss": 0.41408100724220276,
      "eval_runtime": 7.6725,
      "eval_samples_per_second": 488.759,
      "eval_steps_per_second": 7.69,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.004524765536189079,
      "learning_rate": 1.9143099846872597e-05,
      "loss": 0.0122,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9058666666666667,
      "eval_loss": 0.3756696879863739,
      "eval_runtime": 7.8028,
      "eval_samples_per_second": 480.595,
      "eval_steps_per_second": 7.561,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.010448678396642208,
      "learning_rate": 1.7383661915504333e-10,
      "loss": 0.0054,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9085333333333333,
      "eval_loss": 0.3748338520526886,
      "eval_runtime": 7.2573,
      "eval_samples_per_second": 516.724,
      "eval_steps_per_second": 8.13,
      "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
}