trainer_state.json

{
  "best_metric": 0.9544,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0479/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": Infinity,
      "learning_rate": 2.667667667667668e-05,
      "loss": 1.1361,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9365333333333333,
      "eval_loss": 0.2791217863559723,
      "eval_runtime": 6.0786,
      "eval_samples_per_second": 616.922,
      "eval_steps_per_second": 9.706,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 40.377403259277344,
      "learning_rate": 2.3343343343343343e-05,
      "loss": 0.1564,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9509333333333333,
      "eval_loss": 0.18480002880096436,
      "eval_runtime": 6.0733,
      "eval_samples_per_second": 617.461,
      "eval_steps_per_second": 9.715,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 20.99485969543457,
      "learning_rate": 2.001001001001001e-05,
      "loss": 0.0648,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9514666666666667,
      "eval_loss": 0.18248219788074493,
      "eval_runtime": 6.061,
      "eval_samples_per_second": 618.707,
      "eval_steps_per_second": 9.734,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.10036443918943405,
      "learning_rate": 1.6676676676676678e-05,
      "loss": 0.0316,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9490666666666666,
      "eval_loss": 0.1951681226491928,
      "eval_runtime": 6.12,
      "eval_samples_per_second": 612.742,
      "eval_steps_per_second": 9.64,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.28022071719169617,
      "learning_rate": 1.3343343343343344e-05,
      "loss": 0.0159,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.948,
      "eval_loss": 0.20594219863414764,
      "eval_runtime": 6.0555,
      "eval_samples_per_second": 619.27,
      "eval_steps_per_second": 9.743,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.03594016283750534,
      "learning_rate": 1.001001001001001e-05,
      "loss": 0.0092,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9544,
      "eval_loss": 0.1953659951686859,
      "eval_runtime": 5.9907,
      "eval_samples_per_second": 625.975,
      "eval_steps_per_second": 9.849,
      "step": 1998
    }
  ],
  "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": 9.88447923735552e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}