trainer_state.json

{
  "best_metric": 0.9050666666666667,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0841/checkpoints/checkpoint-2331",
  "epoch": 7.0,
  "eval_steps": 500,
  "global_step": 2331,
  "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": 6.655434848731736e-05,
      "loss": 1.7355,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8050666666666667,
      "eval_loss": 0.7174715995788574,
      "eval_runtime": 5.7921,
      "eval_samples_per_second": 647.436,
      "eval_steps_per_second": 10.186,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 2.729964017868042,
      "learning_rate": 5.685900305529986e-05,
      "loss": 0.5238,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8581333333333333,
      "eval_loss": 0.49189072847366333,
      "eval_runtime": 5.9054,
      "eval_samples_per_second": 635.013,
      "eval_steps_per_second": 9.991,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 7.147650718688965,
      "learning_rate": 4.28342139851007e-05,
      "loss": 0.3052,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.876,
      "eval_loss": 0.4096062481403351,
      "eval_runtime": 5.8469,
      "eval_samples_per_second": 641.36,
      "eval_steps_per_second": 10.091,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.2377723604440689,
      "learning_rate": 2.7257762754823102e-05,
      "loss": 0.1832,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8877333333333334,
      "eval_loss": 0.3772968351840973,
      "eval_runtime": 5.9384,
      "eval_samples_per_second": 631.484,
      "eval_steps_per_second": 9.935,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 14.372974395751953,
      "learning_rate": 1.3214756563261342e-05,
      "loss": 0.0989,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9048,
      "eval_loss": 0.3398764729499817,
      "eval_runtime": 6.2042,
      "eval_samples_per_second": 604.429,
      "eval_steps_per_second": 9.51,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 1.0053187608718872,
      "learning_rate": 3.4865850128228307e-06,
      "loss": 0.0542,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9034666666666666,
      "eval_loss": 0.3366783559322357,
      "eval_runtime": 5.8199,
      "eval_samples_per_second": 644.341,
      "eval_steps_per_second": 10.138,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.5932202935218811,
      "learning_rate": 3.178726637742546e-11,
      "loss": 0.0356,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9050666666666667,
      "eval_loss": 0.34085404872894287,
      "eval_runtime": 5.8324,
      "eval_samples_per_second": 642.959,
      "eval_steps_per_second": 10.116,
      "step": 2331
    }
  ],
  "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": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.153189244358144e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}