trainer_state.json

{
  "best_metric": 0.9272,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0526/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": 14.423980712890625,
      "learning_rate": 2.8514533018536286e-05,
      "loss": 1.2044,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9058666666666667,
      "eval_loss": 0.3527013659477234,
      "eval_runtime": 11.9393,
      "eval_samples_per_second": 314.089,
      "eval_steps_per_second": 4.942,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.4579521119594574,
      "learning_rate": 2.4352347027881003e-05,
      "loss": 0.198,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9237333333333333,
      "eval_loss": 0.27222979068756104,
      "eval_runtime": 11.812,
      "eval_samples_per_second": 317.475,
      "eval_steps_per_second": 4.995,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.1366424411535263,
      "learning_rate": 1.8337814009344716e-05,
      "loss": 0.0812,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9226666666666666,
      "eval_loss": 0.2622970640659332,
      "eval_runtime": 11.7836,
      "eval_samples_per_second": 318.239,
      "eval_steps_per_second": 5.007,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.07601583003997803,
      "learning_rate": 1.1662185990655285e-05,
      "loss": 0.0327,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9218666666666666,
      "eval_loss": 0.27649828791618347,
      "eval_runtime": 11.741,
      "eval_samples_per_second": 319.394,
      "eval_steps_per_second": 5.025,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.03594595566391945,
      "learning_rate": 5.647652972118998e-06,
      "loss": 0.0142,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9264,
      "eval_loss": 0.2729930877685547,
      "eval_runtime": 11.707,
      "eval_samples_per_second": 320.321,
      "eval_steps_per_second": 5.04,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.06662894785404205,
      "learning_rate": 1.4854669814637145e-06,
      "loss": 0.0082,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9272,
      "eval_loss": 0.277814120054245,
      "eval_runtime": 12.4742,
      "eval_samples_per_second": 300.621,
      "eval_steps_per_second": 4.73,
      "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
}