trainer_state.json

{
  "best_metric": 0.8682666666666666,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0140/checkpoints/checkpoint-2664",
  "epoch": 8.0,
  "eval_steps": 500,
  "global_step": 2664,
  "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": 0.0002886495269365962,
      "loss": 2.0127,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6888,
      "eval_loss": 1.0778459310531616,
      "eval_runtime": 6.4758,
      "eval_samples_per_second": 579.081,
      "eval_steps_per_second": 9.111,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 0.00025631588426322575,
      "loss": 0.7415,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7650666666666667,
      "eval_loss": 0.7827797532081604,
      "eval_runtime": 6.7966,
      "eval_samples_per_second": 551.745,
      "eval_steps_per_second": 8.681,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 8.46395492553711,
      "learning_rate": 0.0002077292077256867,
      "loss": 0.4367,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8056,
      "eval_loss": 0.6343628764152527,
      "eval_runtime": 6.383,
      "eval_samples_per_second": 587.501,
      "eval_steps_per_second": 9.243,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.19207283854484558,
      "learning_rate": 0.00015035378262848306,
      "loss": 0.2386,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8213333333333334,
      "eval_loss": 0.6032134294509888,
      "eval_runtime": 6.5219,
      "eval_samples_per_second": 574.984,
      "eval_steps_per_second": 9.046,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.16376422345638275,
      "learning_rate": 9.29244973331404e-05,
      "loss": 0.1145,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8485333333333334,
      "eval_loss": 0.5663112998008728,
      "eval_runtime": 6.5806,
      "eval_samples_per_second": 569.854,
      "eval_steps_per_second": 8.966,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.185723215341568,
      "learning_rate": 4.4184439928107014e-05,
      "loss": 0.0427,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8501333333333333,
      "eval_loss": 0.5473652482032776,
      "eval_runtime": 6.2945,
      "eval_samples_per_second": 595.754,
      "eval_steps_per_second": 9.373,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.05701051652431488,
      "learning_rate": 1.1553842323578666e-05,
      "loss": 0.0131,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8634666666666667,
      "eval_loss": 0.5174842476844788,
      "eval_runtime": 6.4142,
      "eval_samples_per_second": 584.639,
      "eval_steps_per_second": 9.198,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 6.468441009521484,
      "learning_rate": 4.172077409214658e-10,
      "loss": 0.0056,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8682666666666666,
      "eval_loss": 0.5099840760231018,
      "eval_runtime": 6.375,
      "eval_samples_per_second": 588.235,
      "eval_steps_per_second": 9.255,
      "step": 2664
    }
  ],
  "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": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.317930564980736e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}