trainer_state.json

{
  "best_metric": 0.8885333333333333,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0570/checkpoints/checkpoint-2997",
  "epoch": 9.0,
  "eval_steps": 500,
  "global_step": 2997,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 198.11215209960938,
      "learning_rate": 9e-05,
      "loss": 2.8632,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7029333333333333,
      "eval_loss": 1.1432005167007446,
      "eval_runtime": 9.2191,
      "eval_samples_per_second": 406.766,
      "eval_steps_per_second": 6.4,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 161.6055908203125,
      "learning_rate": 9e-05,
      "loss": 0.826,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8152,
      "eval_loss": 0.610271155834198,
      "eval_runtime": 9.0503,
      "eval_samples_per_second": 414.353,
      "eval_steps_per_second": 6.519,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 124.0531997680664,
      "learning_rate": 9e-05,
      "loss": 0.528,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8437333333333333,
      "eval_loss": 0.48132163286209106,
      "eval_runtime": 9.5976,
      "eval_samples_per_second": 390.724,
      "eval_steps_per_second": 6.147,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 64.43231201171875,
      "learning_rate": 9e-05,
      "loss": 0.3971,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8565333333333334,
      "eval_loss": 0.4608285129070282,
      "eval_runtime": 9.1374,
      "eval_samples_per_second": 410.401,
      "eval_steps_per_second": 6.457,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 116.30998229980469,
      "learning_rate": 9e-05,
      "loss": 0.316,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8712,
      "eval_loss": 0.4133520722389221,
      "eval_runtime": 9.2563,
      "eval_samples_per_second": 405.131,
      "eval_steps_per_second": 6.374,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 175.6914520263672,
      "learning_rate": 9e-05,
      "loss": 0.253,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8650666666666667,
      "eval_loss": 0.4288407862186432,
      "eval_runtime": 9.3255,
      "eval_samples_per_second": 402.121,
      "eval_steps_per_second": 6.327,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 66.08152770996094,
      "learning_rate": 9e-05,
      "loss": 0.2008,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8773333333333333,
      "eval_loss": 0.4072171747684479,
      "eval_runtime": 9.5612,
      "eval_samples_per_second": 392.21,
      "eval_steps_per_second": 6.171,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 63.93110656738281,
      "learning_rate": 9e-05,
      "loss": 0.1676,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8738666666666667,
      "eval_loss": 0.4257516860961914,
      "eval_runtime": 9.0488,
      "eval_samples_per_second": 414.419,
      "eval_steps_per_second": 6.52,
      "step": 2664
    },
    {
      "epoch": 9.0,
      "grad_norm": 69.3922119140625,
      "learning_rate": 9e-05,
      "loss": 0.1448,
      "step": 2997
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.8885333333333333,
      "eval_loss": 0.39328503608703613,
      "eval_runtime": 9.072,
      "eval_samples_per_second": 413.359,
      "eval_steps_per_second": 6.504,
      "step": 2997
    }
  ],
  "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": true
      },
      "attributes": {}
    }
  },
  "total_flos": 7.3587863155968e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}