trainer_state.json

{
  "best_metric": 0.8936,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0387/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": 2.125685691833496,
      "learning_rate": 6.788924172750679e-05,
      "loss": 1.8606,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7752,
      "eval_loss": 0.8112849593162537,
      "eval_runtime": 6.3991,
      "eval_samples_per_second": 586.024,
      "eval_steps_per_second": 9.22,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 6.183512375367748e-05,
      "loss": 0.5844,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8421333333333333,
      "eval_loss": 0.5220909714698792,
      "eval_runtime": 6.2826,
      "eval_samples_per_second": 596.886,
      "eval_steps_per_second": 9.391,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 3.729097366333008,
      "learning_rate": 5.253176364164983e-05,
      "loss": 0.3567,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8437333333333333,
      "eval_loss": 0.5104581117630005,
      "eval_runtime": 6.4341,
      "eval_samples_per_second": 582.829,
      "eval_steps_per_second": 9.17,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.6496325731277466,
      "learning_rate": 4.1113814093164575e-05,
      "loss": 0.2279,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8605333333333334,
      "eval_loss": 0.4457673132419586,
      "eval_runtime": 6.2453,
      "eval_samples_per_second": 600.452,
      "eval_steps_per_second": 9.447,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.4475836753845215,
      "learning_rate": 2.8958448334757455e-05,
      "loss": 0.1441,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.876,
      "eval_loss": 0.4165574610233307,
      "eval_runtime": 6.2763,
      "eval_samples_per_second": 597.485,
      "eval_steps_per_second": 9.4,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.25858715176582336,
      "learning_rate": 1.7531782870984953e-05,
      "loss": 0.0834,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8842666666666666,
      "eval_loss": 0.40380510687828064,
      "eval_runtime": 6.3235,
      "eval_samples_per_second": 593.025,
      "eval_steps_per_second": 9.33,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 1.5370768308639526,
      "learning_rate": 8.212042196399656e-06,
      "loss": 0.0495,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8861333333333333,
      "eval_loss": 0.40271419286727905,
      "eval_runtime": 6.2593,
      "eval_samples_per_second": 599.106,
      "eval_steps_per_second": 9.426,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.29160067439079285,
      "learning_rate": 2.1233245810799825e-06,
      "loss": 0.0292,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8936,
      "eval_loss": 0.38594380021095276,
      "eval_runtime": 6.3576,
      "eval_samples_per_second": 589.841,
      "eval_steps_per_second": 9.28,
      "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
}