trainer_state.json

{
  "best_metric": 0.8461333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0352/checkpoints/checkpoint-1665",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 1665,
  "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": 9e-05,
      "loss": 2.5604,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7074666666666667,
      "eval_loss": 1.095690131187439,
      "eval_runtime": 6.2761,
      "eval_samples_per_second": 597.509,
      "eval_steps_per_second": 9.401,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 9e-05,
      "loss": 0.7687,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7986666666666666,
      "eval_loss": 0.676312267780304,
      "eval_runtime": 7.0615,
      "eval_samples_per_second": 531.045,
      "eval_steps_per_second": 8.355,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.3384237289428711,
      "learning_rate": 9e-05,
      "loss": 0.4013,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8202666666666667,
      "eval_loss": 0.5954374074935913,
      "eval_runtime": 5.9544,
      "eval_samples_per_second": 629.792,
      "eval_steps_per_second": 9.909,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.5611045360565186,
      "learning_rate": 9e-05,
      "loss": 0.2424,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8341333333333333,
      "eval_loss": 0.5833695530891418,
      "eval_runtime": 5.7852,
      "eval_samples_per_second": 648.21,
      "eval_steps_per_second": 10.199,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.08151687681674957,
      "learning_rate": 9e-05,
      "loss": 0.1642,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8461333333333333,
      "eval_loss": 0.5404489636421204,
      "eval_runtime": 5.7032,
      "eval_samples_per_second": 657.527,
      "eval_steps_per_second": 10.345,
      "step": 1665
    }
  ],
  "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": true
      },
      "attributes": {}
    }
  },
  "total_flos": 8.2370660311296e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}