trainer_state.json

{
  "best_metric": 0.8584,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0022/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": 172.25624084472656,
      "learning_rate": 4.8096988312782174e-05,
      "loss": 3.6153,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5421333333333334,
      "eval_loss": 2.918996810913086,
      "eval_runtime": 7.9478,
      "eval_samples_per_second": 471.83,
      "eval_steps_per_second": 7.423,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 250.7557830810547,
      "learning_rate": 4.267766952966369e-05,
      "loss": 1.617,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7610666666666667,
      "eval_loss": 0.8777552843093872,
      "eval_runtime": 7.7802,
      "eval_samples_per_second": 481.992,
      "eval_steps_per_second": 7.583,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 127.15276336669922,
      "learning_rate": 3.456708580912725e-05,
      "loss": 0.7412,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8082666666666667,
      "eval_loss": 0.6453647017478943,
      "eval_runtime": 8.0132,
      "eval_samples_per_second": 467.976,
      "eval_steps_per_second": 7.363,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 199.5069122314453,
      "learning_rate": 2.5e-05,
      "loss": 0.5529,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8365333333333334,
      "eval_loss": 0.5405584573745728,
      "eval_runtime": 8.0016,
      "eval_samples_per_second": 468.655,
      "eval_steps_per_second": 7.373,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 222.4784393310547,
      "learning_rate": 1.5432914190872757e-05,
      "loss": 0.4693,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8493333333333334,
      "eval_loss": 0.49500513076782227,
      "eval_runtime": 7.8367,
      "eval_samples_per_second": 478.515,
      "eval_steps_per_second": 7.529,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 126.18603515625,
      "learning_rate": 7.3223304703363135e-06,
      "loss": 0.4229,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8517333333333333,
      "eval_loss": 0.4783564507961273,
      "eval_runtime": 7.3482,
      "eval_samples_per_second": 510.331,
      "eval_steps_per_second": 8.029,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 306.167236328125,
      "learning_rate": 1.9030116872178316e-06,
      "loss": 0.4021,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8568,
      "eval_loss": 0.4630471467971802,
      "eval_runtime": 7.4171,
      "eval_samples_per_second": 505.591,
      "eval_steps_per_second": 7.955,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 693.2731323242188,
      "learning_rate": 0.0,
      "loss": 0.3882,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8584,
      "eval_loss": 0.46055299043655396,
      "eval_runtime": 8.162,
      "eval_samples_per_second": 459.447,
      "eval_steps_per_second": 7.229,
      "step": 2664
    }
  ],
  "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": true
      },
      "attributes": {}
    }
  },
  "total_flos": 6.5411433916416e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}