trainer_state.json

{
  "best_metric": 0.9173333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0069/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": 29.722087860107422,
      "learning_rate": 8.659485808097889e-05,
      "loss": 1.3548,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8090666666666667,
      "eval_loss": 0.673751711845398,
      "eval_runtime": 6.8063,
      "eval_samples_per_second": 550.963,
      "eval_steps_per_second": 8.668,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 3.173098564147949,
      "learning_rate": 7.685730736804761e-05,
      "loss": 0.4193,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8533333333333334,
      "eval_loss": 0.5064404010772705,
      "eval_runtime": 6.546,
      "eval_samples_per_second": 572.865,
      "eval_steps_per_second": 9.013,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": Infinity,
      "learning_rate": 6.231876231770601e-05,
      "loss": 0.2457,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8837333333333334,
      "eval_loss": 0.3840847313404083,
      "eval_runtime": 6.6954,
      "eval_samples_per_second": 560.084,
      "eval_steps_per_second": 8.812,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.5161293745040894,
      "learning_rate": 4.510613478854492e-05,
      "loss": 0.1351,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8978666666666667,
      "eval_loss": 0.36032167077064514,
      "eval_runtime": 6.5246,
      "eval_samples_per_second": 574.748,
      "eval_steps_per_second": 9.043,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 22.62803840637207,
      "learning_rate": 2.7877349199942123e-05,
      "loss": 0.0672,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9048,
      "eval_loss": 0.35179126262664795,
      "eval_runtime": 6.4873,
      "eval_samples_per_second": 578.051,
      "eval_steps_per_second": 9.095,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 1.9570411443710327,
      "learning_rate": 1.3255331978432106e-05,
      "loss": 0.0263,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9101333333333333,
      "eval_loss": 0.3458578884601593,
      "eval_runtime": 6.4951,
      "eval_samples_per_second": 577.357,
      "eval_steps_per_second": 9.084,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.15059025585651398,
      "learning_rate": 3.4661526970736e-06,
      "loss": 0.0108,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9173333333333333,
      "eval_loss": 0.3246249258518219,
      "eval_runtime": 6.471,
      "eval_samples_per_second": 579.505,
      "eval_steps_per_second": 9.118,
      "step": 2331
    }
  ],
  "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": 1.153189244358144e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}