trainer_state.json

{
  "best_metric": 0.9392,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0578/checkpoints/checkpoint-1998",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 1998,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 0.6746072173118591,
      "learning_rate": 3e-05,
      "loss": 1.166,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9274666666666667,
      "eval_loss": 0.3021855652332306,
      "eval_runtime": 12.3887,
      "eval_samples_per_second": 302.696,
      "eval_steps_per_second": 4.762,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.2053106278181076,
      "learning_rate": 3e-05,
      "loss": 0.1787,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9322666666666667,
      "eval_loss": 0.24098815023899078,
      "eval_runtime": 11.8847,
      "eval_samples_per_second": 315.531,
      "eval_steps_per_second": 4.964,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.06273005157709122,
      "learning_rate": 3e-05,
      "loss": 0.0842,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.932,
      "eval_loss": 0.2299923151731491,
      "eval_runtime": 12.6959,
      "eval_samples_per_second": 295.372,
      "eval_steps_per_second": 4.647,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.07098501175642014,
      "learning_rate": 3e-05,
      "loss": 0.0438,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9341333333333334,
      "eval_loss": 0.24950742721557617,
      "eval_runtime": 12.4915,
      "eval_samples_per_second": 300.205,
      "eval_steps_per_second": 4.723,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.0565139539539814,
      "learning_rate": 3e-05,
      "loss": 0.0258,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9352,
      "eval_loss": 0.2574763298034668,
      "eval_runtime": 12.7223,
      "eval_samples_per_second": 294.757,
      "eval_steps_per_second": 4.638,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.28586670756340027,
      "learning_rate": 3e-05,
      "loss": 0.018,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9392,
      "eval_loss": 0.2564932703971863,
      "eval_runtime": 12.1195,
      "eval_samples_per_second": 309.419,
      "eval_steps_per_second": 4.868,
      "step": 1998
    }
  ],
  "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": 9.88447923735552e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}