trainer_state.json

{
  "best_metric": 0.9416,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0280/checkpoints/checkpoint-1332",
  "epoch": 4.0,
  "eval_steps": 500,
  "global_step": 1332,
  "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": 7.205405405405405e-05,
      "loss": 0.6251,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9189333333333334,
      "eval_loss": 0.2511806786060333,
      "eval_runtime": 6.1902,
      "eval_samples_per_second": 605.795,
      "eval_steps_per_second": 9.531,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 4.8375678062438965,
      "learning_rate": 5.405405405405406e-05,
      "loss": 0.1227,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9256,
      "eval_loss": 0.23581796884536743,
      "eval_runtime": 6.275,
      "eval_samples_per_second": 597.605,
      "eval_steps_per_second": 9.402,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.027780085802078247,
      "learning_rate": 3.6054054054054056e-05,
      "loss": 0.0371,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9344,
      "eval_loss": 0.22402900457382202,
      "eval_runtime": 6.3183,
      "eval_samples_per_second": 593.516,
      "eval_steps_per_second": 9.338,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.15983650088310242,
      "learning_rate": 1.8054054054054053e-05,
      "loss": 0.0089,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9416,
      "eval_loss": 0.22367602586746216,
      "eval_runtime": 7.0687,
      "eval_samples_per_second": 530.508,
      "eval_steps_per_second": 8.347,
      "step": 1332
    }
  ],
  "logging_steps": 500,
  "max_steps": 1665,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "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": 6.58965282490368e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}