trainer_state.json

{
  "best_metric": 0.8626666666666667,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0652/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": 7e-05,
      "loss": 1.6259,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.816,
      "eval_loss": 0.6959978938102722,
      "eval_runtime": 4.7898,
      "eval_samples_per_second": 782.915,
      "eval_steps_per_second": 12.318,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 2.527723550796509,
      "learning_rate": 7e-05,
      "loss": 0.4876,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8114666666666667,
      "eval_loss": 0.6173341274261475,
      "eval_runtime": 4.1612,
      "eval_samples_per_second": 901.178,
      "eval_steps_per_second": 14.179,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 2.583649158477783,
      "learning_rate": 7e-05,
      "loss": 0.2984,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8405333333333334,
      "eval_loss": 0.5540310144424438,
      "eval_runtime": 4.1265,
      "eval_samples_per_second": 908.753,
      "eval_steps_per_second": 14.298,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": Infinity,
      "learning_rate": 7e-05,
      "loss": 0.2049,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8602666666666666,
      "eval_loss": 0.49156665802001953,
      "eval_runtime": 4.1287,
      "eval_samples_per_second": 908.286,
      "eval_steps_per_second": 14.29,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 1.8416281938552856,
      "learning_rate": 7e-05,
      "loss": 0.1509,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8626666666666667,
      "eval_loss": 0.4845036566257477,
      "eval_runtime": 4.0463,
      "eval_samples_per_second": 926.768,
      "eval_steps_per_second": 14.581,
      "step": 1665
    }
  ],
  "logging_steps": 500,
  "max_steps": 2664,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "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": 8.2370660311296e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}