trainer_state.json

{
  "best_metric": 0.8693333333333333,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0226/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": 130.7045135498047,
      "learning_rate": 3e-05,
      "loss": 3.823,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.49493333333333334,
      "eval_loss": 3.625760078430176,
      "eval_runtime": 5.1023,
      "eval_samples_per_second": 734.965,
      "eval_steps_per_second": 11.563,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 292.2391662597656,
      "learning_rate": 3e-05,
      "loss": 2.9708,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6429333333333334,
      "eval_loss": 2.050452709197998,
      "eval_runtime": 5.0801,
      "eval_samples_per_second": 738.172,
      "eval_steps_per_second": 11.614,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 123.69570922851562,
      "learning_rate": 3e-05,
      "loss": 1.3072,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7618666666666667,
      "eval_loss": 0.9252050518989563,
      "eval_runtime": 5.0657,
      "eval_samples_per_second": 740.268,
      "eval_steps_per_second": 11.647,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 83.30982971191406,
      "learning_rate": 3e-05,
      "loss": 0.7789,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8125333333333333,
      "eval_loss": 0.6645534038543701,
      "eval_runtime": 5.0715,
      "eval_samples_per_second": 739.423,
      "eval_steps_per_second": 11.634,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 196.97056579589844,
      "learning_rate": 3e-05,
      "loss": 0.5914,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8333333333333334,
      "eval_loss": 0.5627994537353516,
      "eval_runtime": 5.0351,
      "eval_samples_per_second": 744.768,
      "eval_steps_per_second": 11.718,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 349.22259521484375,
      "learning_rate": 3e-05,
      "loss": 0.4808,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.852,
      "eval_loss": 0.49497321248054504,
      "eval_runtime": 5.1852,
      "eval_samples_per_second": 723.21,
      "eval_steps_per_second": 11.378,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 131.05450439453125,
      "learning_rate": 3e-05,
      "loss": 0.3974,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8653333333333333,
      "eval_loss": 0.45529961585998535,
      "eval_runtime": 5.0345,
      "eval_samples_per_second": 744.863,
      "eval_steps_per_second": 11.719,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 116.65331268310547,
      "learning_rate": 3e-05,
      "loss": 0.3325,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8693333333333333,
      "eval_loss": 0.4422469735145569,
      "eval_runtime": 5.0737,
      "eval_samples_per_second": 739.112,
      "eval_steps_per_second": 11.629,
      "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
}