trainer_state.json

{
  "best_metric": 0.944,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0664/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": 3e-05,
      "loss": 1.1092,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9410666666666667,
      "eval_loss": 0.254812091588974,
      "eval_runtime": 12.4098,
      "eval_samples_per_second": 302.182,
      "eval_steps_per_second": 4.754,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.24436338245868683,
      "learning_rate": 3e-05,
      "loss": 0.1538,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9394666666666667,
      "eval_loss": 0.2029528021812439,
      "eval_runtime": 12.3324,
      "eval_samples_per_second": 304.076,
      "eval_steps_per_second": 4.784,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 9.051680564880371,
      "learning_rate": 3e-05,
      "loss": 0.0673,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9365333333333333,
      "eval_loss": 0.22422005236148834,
      "eval_runtime": 12.3244,
      "eval_samples_per_second": 304.275,
      "eval_steps_per_second": 4.787,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.16816633939743042,
      "learning_rate": 3e-05,
      "loss": 0.0332,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9392,
      "eval_loss": 0.2259085774421692,
      "eval_runtime": 12.2095,
      "eval_samples_per_second": 307.139,
      "eval_steps_per_second": 4.832,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.03614787012338638,
      "learning_rate": 3e-05,
      "loss": 0.0163,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9429333333333333,
      "eval_loss": 0.213191956281662,
      "eval_runtime": 12.2858,
      "eval_samples_per_second": 305.229,
      "eval_steps_per_second": 4.802,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.0286348145455122,
      "learning_rate": 3e-05,
      "loss": 0.0097,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9429333333333333,
      "eval_loss": 0.22389760613441467,
      "eval_runtime": 12.4729,
      "eval_samples_per_second": 300.653,
      "eval_steps_per_second": 4.73,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.04926493391394615,
      "learning_rate": 3e-05,
      "loss": 0.0094,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.944,
      "eval_loss": 0.24098949134349823,
      "eval_runtime": 12.4742,
      "eval_samples_per_second": 300.621,
      "eval_steps_per_second": 4.73,
      "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
}