trainer_state.json

{
  "best_metric": 0.9474666666666667,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0729/checkpoints/checkpoint-1998",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 1998,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 0.8878992199897766,
      "learning_rate": 6.653391037658466e-05,
      "loss": 0.7515,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9290666666666667,
      "eval_loss": 0.25241854786872864,
      "eval_runtime": 13.1143,
      "eval_samples_per_second": 285.947,
      "eval_steps_per_second": 4.499,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.07038387656211853,
      "learning_rate": 5.682214306505567e-05,
      "loss": 0.1331,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.92,
      "eval_loss": 0.2850964665412903,
      "eval_runtime": 12.5883,
      "eval_samples_per_second": 297.896,
      "eval_steps_per_second": 4.687,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.06545305252075195,
      "learning_rate": 4.2788232688471e-05,
      "loss": 0.0485,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9301333333333334,
      "eval_loss": 0.2703464925289154,
      "eval_runtime": 12.5251,
      "eval_samples_per_second": 299.4,
      "eval_steps_per_second": 4.711,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.02720840647816658,
      "learning_rate": 2.7211767311528996e-05,
      "loss": 0.0157,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9437333333333333,
      "eval_loss": 0.22763261198997498,
      "eval_runtime": 12.4193,
      "eval_samples_per_second": 301.95,
      "eval_steps_per_second": 4.751,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.019502338021993637,
      "learning_rate": 1.3177856934944328e-05,
      "loss": 0.0061,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9437333333333333,
      "eval_loss": 0.2280394434928894,
      "eval_runtime": 12.3612,
      "eval_samples_per_second": 303.37,
      "eval_steps_per_second": 4.773,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.07268548756837845,
      "learning_rate": 3.4660896234153336e-06,
      "loss": 0.0036,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9474666666666667,
      "eval_loss": 0.223210871219635,
      "eval_runtime": 12.308,
      "eval_samples_per_second": 304.681,
      "eval_steps_per_second": 4.794,
      "step": 1998
    }
  ],
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 9.88447923735552e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}