trainer_state.json

{
  "best_metric": 0.5237333333333334,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0714/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": 0.0005,
      "loss": 3.1549,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.2464,
      "eval_loss": 2.7680068016052246,
      "eval_runtime": 10.3044,
      "eval_samples_per_second": 363.921,
      "eval_steps_per_second": 5.726,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 0.0005,
      "loss": 2.5534,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.3408,
      "eval_loss": 2.3918583393096924,
      "eval_runtime": 10.3088,
      "eval_samples_per_second": 363.768,
      "eval_steps_per_second": 5.723,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 11.021343231201172,
      "learning_rate": 0.0005,
      "loss": 2.1429,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.3936,
      "eval_loss": 2.1302378177642822,
      "eval_runtime": 10.3885,
      "eval_samples_per_second": 360.976,
      "eval_steps_per_second": 5.679,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 14.008143424987793,
      "learning_rate": 0.0005,
      "loss": 1.8449,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.43733333333333335,
      "eval_loss": 1.9548763036727905,
      "eval_runtime": 11.0803,
      "eval_samples_per_second": 338.438,
      "eval_steps_per_second": 5.325,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 11.996915817260742,
      "learning_rate": 0.0005,
      "loss": 1.5834,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.47573333333333334,
      "eval_loss": 1.8434683084487915,
      "eval_runtime": 10.4024,
      "eval_samples_per_second": 360.494,
      "eval_steps_per_second": 5.672,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 17.0865421295166,
      "learning_rate": 0.0005,
      "loss": 1.3287,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.48906666666666665,
      "eval_loss": 1.8183892965316772,
      "eval_runtime": 10.2285,
      "eval_samples_per_second": 366.624,
      "eval_steps_per_second": 5.768,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 5.588679790496826,
      "learning_rate": 0.0005,
      "loss": 1.0328,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.5237333333333334,
      "eval_loss": 1.7224879264831543,
      "eval_runtime": 10.1957,
      "eval_samples_per_second": 367.801,
      "eval_steps_per_second": 5.787,
      "step": 2331
    }
  ],
  "logging_steps": 500,
  "max_steps": 2997,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9,
  "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": 1.153189244358144e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}