trainer_state.json

{
  "best_metric": 0.8936,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0914/checkpoints/checkpoint-2331",
  "epoch": 7.0,
  "eval_steps": 500,
  "global_step": 2331,
  "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": 8.575718575718577e-05,
      "loss": 1.882,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7666666666666667,
      "eval_loss": 0.8143971562385559,
      "eval_runtime": 7.4108,
      "eval_samples_per_second": 506.019,
      "eval_steps_per_second": 7.961,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 11.984903335571289,
      "learning_rate": 7.147147147147148e-05,
      "loss": 0.5885,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8272,
      "eval_loss": 0.5748838782310486,
      "eval_runtime": 6.8316,
      "eval_samples_per_second": 548.919,
      "eval_steps_per_second": 8.636,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": Infinity,
      "learning_rate": 5.7228657228657234e-05,
      "loss": 0.3248,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8592,
      "eval_loss": 0.4654926359653473,
      "eval_runtime": 7.0495,
      "eval_samples_per_second": 531.951,
      "eval_steps_per_second": 8.369,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.6059653162956238,
      "learning_rate": 4.294294294294294e-05,
      "loss": 0.1787,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8672,
      "eval_loss": 0.43710747361183167,
      "eval_runtime": 6.8399,
      "eval_samples_per_second": 548.251,
      "eval_steps_per_second": 8.626,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.7784273028373718,
      "learning_rate": 2.8657228657228656e-05,
      "loss": 0.096,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8752,
      "eval_loss": 0.4238051474094391,
      "eval_runtime": 6.7435,
      "eval_samples_per_second": 556.087,
      "eval_steps_per_second": 8.749,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.05428040400147438,
      "learning_rate": 1.4371514371514371e-05,
      "loss": 0.0467,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8877333333333334,
      "eval_loss": 0.40625742077827454,
      "eval_runtime": 7.4882,
      "eval_samples_per_second": 500.786,
      "eval_steps_per_second": 7.879,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.04293455183506012,
      "learning_rate": 8.580008580008581e-08,
      "loss": 0.0198,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8936,
      "eval_loss": 0.3967723250389099,
      "eval_runtime": 6.8316,
      "eval_samples_per_second": 548.918,
      "eval_steps_per_second": 8.636,
      "step": 2331
    }
  ],
  "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": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.153189244358144e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}