trainer_state.json

{
  "best_metric": 0.9416,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0909/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": 8.892225558892226e-05,
      "loss": 0.6479,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.92,
      "eval_loss": 0.2617263197898865,
      "eval_runtime": 6.4628,
      "eval_samples_per_second": 580.247,
      "eval_steps_per_second": 9.129,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.8425988554954529,
      "learning_rate": 7.781114447781115e-05,
      "loss": 0.16,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9178666666666667,
      "eval_loss": 0.28157591819763184,
      "eval_runtime": 6.5037,
      "eval_samples_per_second": 576.595,
      "eval_steps_per_second": 9.072,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 2.1107635498046875,
      "learning_rate": 6.670003336670004e-05,
      "loss": 0.0749,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9224,
      "eval_loss": 0.2684481143951416,
      "eval_runtime": 6.5257,
      "eval_samples_per_second": 574.649,
      "eval_steps_per_second": 9.041,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.10983747988939285,
      "learning_rate": 5.5588922255588925e-05,
      "loss": 0.0342,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9226666666666666,
      "eval_loss": 0.2849978804588318,
      "eval_runtime": 6.4736,
      "eval_samples_per_second": 579.278,
      "eval_steps_per_second": 9.114,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.11726373434066772,
      "learning_rate": 4.447781114447781e-05,
      "loss": 0.0147,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9365333333333333,
      "eval_loss": 0.2725699245929718,
      "eval_runtime": 6.5082,
      "eval_samples_per_second": 576.197,
      "eval_steps_per_second": 9.065,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.07535902410745621,
      "learning_rate": 3.33667000333667e-05,
      "loss": 0.0065,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.94,
      "eval_loss": 0.26398804783821106,
      "eval_runtime": 6.4578,
      "eval_samples_per_second": 580.695,
      "eval_steps_per_second": 9.136,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.09727034717798233,
      "learning_rate": 2.225558892225559e-05,
      "loss": 0.0034,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9378666666666666,
      "eval_loss": 0.2772713005542755,
      "eval_runtime": 6.48,
      "eval_samples_per_second": 578.701,
      "eval_steps_per_second": 9.105,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.011889305897057056,
      "learning_rate": 1.1144477811144478e-05,
      "loss": 0.0022,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9416,
      "eval_loss": 0.2661619484424591,
      "eval_runtime": 6.2857,
      "eval_samples_per_second": 596.594,
      "eval_steps_per_second": 9.386,
      "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
}