trainer_state.json

{
  "best_metric": 0.8885333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0278/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": 4.525311529382417e-05,
      "loss": 2.0577,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7754666666666666,
      "eval_loss": 0.9113693237304688,
      "eval_runtime": 11.0709,
      "eval_samples_per_second": 338.725,
      "eval_steps_per_second": 5.329,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 3.281509432839476e-05,
      "loss": 0.601,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8509333333333333,
      "eval_loss": 0.5507181286811829,
      "eval_runtime": 11.2866,
      "eval_samples_per_second": 332.254,
      "eval_steps_per_second": 5.227,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 2.4969398975372314,
      "learning_rate": 1.736435462495525e-05,
      "loss": 0.2968,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8693333333333333,
      "eval_loss": 0.4559718072414398,
      "eval_runtime": 10.9196,
      "eval_samples_per_second": 343.419,
      "eval_steps_per_second": 5.403,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.7973734736442566,
      "learning_rate": 4.8301719287419e-06,
      "loss": 0.1513,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8848,
      "eval_loss": 0.4051165282726288,
      "eval_runtime": 10.7983,
      "eval_samples_per_second": 347.278,
      "eval_steps_per_second": 5.464,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 1.680179238319397,
      "learning_rate": 1.7800850740101738e-10,
      "loss": 0.092,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8885333333333333,
      "eval_loss": 0.3876320421695709,
      "eval_runtime": 10.7876,
      "eval_samples_per_second": 347.621,
      "eval_steps_per_second": 5.469,
      "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
}