trainer_state.json

{
  "best_metric": 0.9522666666666667,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0213/checkpoints/checkpoint-2331",
  "epoch": 7.0,
  "eval_steps": 500,
  "global_step": 2331,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 2.6976161003112793,
      "learning_rate": 8.554359905560886e-05,
      "loss": 0.6275,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9285333333333333,
      "eval_loss": 0.24837812781333923,
      "eval_runtime": 7.2913,
      "eval_samples_per_second": 514.311,
      "eval_steps_per_second": 8.092,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.028960146009922028,
      "learning_rate": 7.305704108364301e-05,
      "loss": 0.1188,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9357333333333333,
      "eval_loss": 0.2327362298965454,
      "eval_runtime": 7.0246,
      "eval_samples_per_second": 533.836,
      "eval_steps_per_second": 8.399,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.011435726657509804,
      "learning_rate": 5.501344202803415e-05,
      "loss": 0.0423,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9432,
      "eval_loss": 0.22233636677265167,
      "eval_runtime": 6.7185,
      "eval_samples_per_second": 558.16,
      "eval_steps_per_second": 8.782,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.02038324438035488,
      "learning_rate": 3.498655797196586e-05,
      "loss": 0.0134,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9501333333333334,
      "eval_loss": 0.216081440448761,
      "eval_runtime": 6.5181,
      "eval_samples_per_second": 575.321,
      "eval_steps_per_second": 9.052,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.0188947431743145,
      "learning_rate": 1.6942958916356995e-05,
      "loss": 0.0042,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9514666666666667,
      "eval_loss": 0.21176356077194214,
      "eval_runtime": 6.4688,
      "eval_samples_per_second": 579.703,
      "eval_steps_per_second": 9.121,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.016306424513459206,
      "learning_rate": 4.456400944391144e-06,
      "loss": 0.0017,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9517333333333333,
      "eval_loss": 0.21258734166622162,
      "eval_runtime": 6.5668,
      "eval_samples_per_second": 571.055,
      "eval_steps_per_second": 8.985,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": NaN,
      "learning_rate": 4.086934248526131e-11,
      "loss": 0.0014,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9522666666666667,
      "eval_loss": 0.2067682445049286,
      "eval_runtime": 6.6098,
      "eval_samples_per_second": 567.338,
      "eval_steps_per_second": 8.926,
      "step": 2331
    }
  ],
  "logging_steps": 500,
  "max_steps": 2331,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "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": 1.153189244358144e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}