trainer_state.json

{
  "best_metric": 0.8981333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0623/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": 8.659485808097889e-05,
      "loss": 1.7359,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.744,
      "eval_loss": 0.833530068397522,
      "eval_runtime": 7.4714,
      "eval_samples_per_second": 501.912,
      "eval_steps_per_second": 7.897,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 8.248693466186523,
      "learning_rate": 7.685730736804761e-05,
      "loss": 0.5936,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8317333333333333,
      "eval_loss": 0.5626587867736816,
      "eval_runtime": 8.0835,
      "eval_samples_per_second": 463.907,
      "eval_steps_per_second": 7.299,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 13.314876556396484,
      "learning_rate": 6.226977039554153e-05,
      "loss": 0.3526,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8429333333333333,
      "eval_loss": 0.4905378222465515,
      "eval_runtime": 7.9999,
      "eval_samples_per_second": 468.756,
      "eval_steps_per_second": 7.375,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.10626498609781265,
      "learning_rate": 4.505306743117272e-05,
      "loss": 0.2139,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8677333333333334,
      "eval_loss": 0.4311425983905792,
      "eval_runtime": 7.3091,
      "eval_samples_per_second": 513.061,
      "eval_steps_per_second": 8.072,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.056557364761829376,
      "learning_rate": 2.7828285431465317e-05,
      "loss": 0.121,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8802666666666666,
      "eval_loss": 0.4112357497215271,
      "eval_runtime": 7.3075,
      "eval_samples_per_second": 513.175,
      "eval_steps_per_second": 8.074,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.3472993075847626,
      "learning_rate": 1.3217741312837138e-05,
      "loss": 0.0619,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8906666666666667,
      "eval_loss": 0.37756314873695374,
      "eval_runtime": 8.3915,
      "eval_samples_per_second": 446.881,
      "eval_steps_per_second": 7.031,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.6734321117401123,
      "learning_rate": 3.4457579724370674e-06,
      "loss": 0.0302,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8922666666666667,
      "eval_loss": 0.3670755922794342,
      "eval_runtime": 7.3996,
      "eval_samples_per_second": 506.783,
      "eval_steps_per_second": 7.973,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.1527414470911026,
      "learning_rate": 3.12905914479078e-11,
      "loss": 0.0192,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8981333333333333,
      "eval_loss": 0.3504849672317505,
      "eval_runtime": 7.3009,
      "eval_samples_per_second": 513.638,
      "eval_steps_per_second": 8.081,
      "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
}