trainer_state.json

{
  "best_metric": 0.9085333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0912/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": Infinity,
      "learning_rate": 6.655434848731736e-05,
      "loss": 1.5439,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8173333333333334,
      "eval_loss": 0.654940128326416,
      "eval_runtime": 5.9179,
      "eval_samples_per_second": 633.67,
      "eval_steps_per_second": 9.97,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.29835620522499084,
      "learning_rate": 5.685900305529986e-05,
      "loss": 0.4275,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8528,
      "eval_loss": 0.5048868656158447,
      "eval_runtime": 5.962,
      "eval_samples_per_second": 628.987,
      "eval_steps_per_second": 9.896,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.23752492666244507,
      "learning_rate": 4.28342139851007e-05,
      "loss": 0.2299,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8789333333333333,
      "eval_loss": 0.4043629467487335,
      "eval_runtime": 6.4261,
      "eval_samples_per_second": 583.553,
      "eval_steps_per_second": 9.181,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": Infinity,
      "learning_rate": 2.7303772261234784e-05,
      "loss": 0.1292,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.892,
      "eval_loss": 0.37236514687538147,
      "eval_runtime": 5.939,
      "eval_samples_per_second": 631.417,
      "eval_steps_per_second": 9.934,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 1.428711175918579,
      "learning_rate": 1.325169576262615e-05,
      "loss": 0.0569,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9029333333333334,
      "eval_loss": 0.34200355410575867,
      "eval_runtime": 5.995,
      "eval_samples_per_second": 625.519,
      "eval_steps_per_second": 9.841,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.6435881853103638,
      "learning_rate": 3.5071376436769933e-06,
      "loss": 0.0251,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9085333333333333,
      "eval_loss": 0.33061182498931885,
      "eval_runtime": 5.8327,
      "eval_samples_per_second": 642.925,
      "eval_steps_per_second": 10.115,
      "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
}