trainer_state.json

{
  "best_metric": 0.8834666666666666,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0821/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": 2.3950605392456055,
      "learning_rate": 5e-05,
      "loss": 1.7444,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8106666666666666,
      "eval_loss": 0.7417201399803162,
      "eval_runtime": 12.9431,
      "eval_samples_per_second": 289.73,
      "eval_steps_per_second": 4.558,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 5e-05,
      "loss": 0.5118,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8442666666666667,
      "eval_loss": 0.5282045602798462,
      "eval_runtime": 11.7803,
      "eval_samples_per_second": 318.328,
      "eval_steps_per_second": 5.008,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.6271276473999023,
      "learning_rate": 5e-05,
      "loss": 0.2655,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8690666666666667,
      "eval_loss": 0.4473636746406555,
      "eval_runtime": 12.4492,
      "eval_samples_per_second": 301.225,
      "eval_steps_per_second": 4.739,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.5355790257453918,
      "learning_rate": 5e-05,
      "loss": 0.1491,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8674666666666667,
      "eval_loss": 0.4574439525604248,
      "eval_runtime": 12.5075,
      "eval_samples_per_second": 299.82,
      "eval_steps_per_second": 4.717,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.15985365211963654,
      "learning_rate": 5e-05,
      "loss": 0.0926,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8794666666666666,
      "eval_loss": 0.4245550036430359,
      "eval_runtime": 11.9649,
      "eval_samples_per_second": 313.417,
      "eval_steps_per_second": 4.931,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.2567598819732666,
      "learning_rate": 5e-05,
      "loss": 0.0622,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8834666666666666,
      "eval_loss": 0.45469290018081665,
      "eval_runtime": 11.8598,
      "eval_samples_per_second": 316.195,
      "eval_steps_per_second": 4.975,
      "step": 1998
    }
  ],
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 9.88447923735552e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}