trainer_state.json

{
  "best_metric": 0.956,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0214/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": 1.4934918880462646,
      "learning_rate": 8.554359905560886e-05,
      "loss": 0.6298,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9330666666666667,
      "eval_loss": 0.24191591143608093,
      "eval_runtime": 6.5986,
      "eval_samples_per_second": 568.302,
      "eval_steps_per_second": 8.941,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 7.310443249967126e-05,
      "loss": 0.1189,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9402666666666667,
      "eval_loss": 0.22558879852294922,
      "eval_runtime": 6.5414,
      "eval_samples_per_second": 573.27,
      "eval_steps_per_second": 9.019,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.052664946764707565,
      "learning_rate": 5.5072560837986624e-05,
      "loss": 0.0428,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9522666666666667,
      "eval_loss": 0.1974857598543167,
      "eval_runtime": 7.5196,
      "eval_samples_per_second": 498.7,
      "eval_steps_per_second": 7.846,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.06115534529089928,
      "learning_rate": 3.504569497048685e-05,
      "loss": 0.0104,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9549333333333333,
      "eval_loss": 0.189637690782547,
      "eval_runtime": 6.7487,
      "eval_samples_per_second": 555.664,
      "eval_steps_per_second": 8.742,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.018230965360999107,
      "learning_rate": 1.6990401295621727e-05,
      "loss": 0.0032,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9544,
      "eval_loss": 0.18555068969726562,
      "eval_runtime": 7.6172,
      "eval_samples_per_second": 492.306,
      "eval_steps_per_second": 7.746,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.05719016492366791,
      "learning_rate": 4.48275215934364e-06,
      "loss": 0.0018,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.956,
      "eval_loss": 0.18424057960510254,
      "eval_runtime": 7.3653,
      "eval_samples_per_second": 509.144,
      "eval_steps_per_second": 8.011,
      "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
}