trainer_state.json

{
  "best_metric": 0.9042666666666667,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0529/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": 4.446112779446113e-05,
      "loss": 1.8953,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7981333333333334,
      "eval_loss": 0.806739330291748,
      "eval_runtime": 6.9701,
      "eval_samples_per_second": 538.013,
      "eval_steps_per_second": 8.465,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 3.892225558892226e-05,
      "loss": 0.5642,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8501333333333333,
      "eval_loss": 0.5320011973381042,
      "eval_runtime": 6.3111,
      "eval_samples_per_second": 594.196,
      "eval_steps_per_second": 9.349,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.4671281576156616,
      "learning_rate": 3.33667000333667e-05,
      "loss": 0.3016,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8773333333333333,
      "eval_loss": 0.4300127923488617,
      "eval_runtime": 6.3886,
      "eval_samples_per_second": 586.98,
      "eval_steps_per_second": 9.235,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 4.932580947875977,
      "learning_rate": 2.7811144477811146e-05,
      "loss": 0.1818,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8861333333333333,
      "eval_loss": 0.40388500690460205,
      "eval_runtime": 6.6427,
      "eval_samples_per_second": 564.533,
      "eval_steps_per_second": 8.882,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 16.322803497314453,
      "learning_rate": 2.225558892225559e-05,
      "loss": 0.1122,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8936,
      "eval_loss": 0.3754415810108185,
      "eval_runtime": 5.9461,
      "eval_samples_per_second": 630.666,
      "eval_steps_per_second": 9.922,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.14778943359851837,
      "learning_rate": 1.6700033366700034e-05,
      "loss": 0.061,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8912,
      "eval_loss": 0.39013683795928955,
      "eval_runtime": 5.951,
      "eval_samples_per_second": 630.148,
      "eval_steps_per_second": 9.914,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.0757766142487526,
      "learning_rate": 1.1144477811144478e-05,
      "loss": 0.0346,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8992,
      "eval_loss": 0.4033607542514801,
      "eval_runtime": 6.3466,
      "eval_samples_per_second": 590.868,
      "eval_steps_per_second": 9.296,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.12226133048534393,
      "learning_rate": 5.5889222555889224e-06,
      "loss": 0.0189,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9042666666666667,
      "eval_loss": 0.382087767124176,
      "eval_runtime": 5.8875,
      "eval_samples_per_second": 636.939,
      "eval_steps_per_second": 10.021,
      "step": 2664
    }
  ],
  "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": 1.317930564980736e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}