trainer_state.json

{
  "best_metric": 0.876,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0356/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": 347.983154296875,
      "learning_rate": 9e-05,
      "loss": 3.0278,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6642666666666667,
      "eval_loss": 1.2886346578598022,
      "eval_runtime": 14.589,
      "eval_samples_per_second": 257.043,
      "eval_steps_per_second": 4.044,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 140.2672576904297,
      "learning_rate": 9e-05,
      "loss": 0.8596,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8152,
      "eval_loss": 0.6093230247497559,
      "eval_runtime": 14.5056,
      "eval_samples_per_second": 258.521,
      "eval_steps_per_second": 4.067,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 57.97684860229492,
      "learning_rate": 9e-05,
      "loss": 0.5419,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8458666666666667,
      "eval_loss": 0.5221971869468689,
      "eval_runtime": 14.5877,
      "eval_samples_per_second": 257.066,
      "eval_steps_per_second": 4.044,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 57.41337585449219,
      "learning_rate": 9e-05,
      "loss": 0.4078,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8602666666666666,
      "eval_loss": 0.46603602170944214,
      "eval_runtime": 14.5046,
      "eval_samples_per_second": 258.539,
      "eval_steps_per_second": 4.068,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 72.65909576416016,
      "learning_rate": 9e-05,
      "loss": 0.3086,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8610666666666666,
      "eval_loss": 0.458729088306427,
      "eval_runtime": 14.7578,
      "eval_samples_per_second": 254.103,
      "eval_steps_per_second": 3.998,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 41.720890045166016,
      "learning_rate": 9e-05,
      "loss": 0.2511,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8738666666666667,
      "eval_loss": 0.42264336347579956,
      "eval_runtime": 14.5336,
      "eval_samples_per_second": 258.023,
      "eval_steps_per_second": 4.06,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 34.57700729370117,
      "learning_rate": 9e-05,
      "loss": 0.2044,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.876,
      "eval_loss": 0.41637590527534485,
      "eval_runtime": 14.5437,
      "eval_samples_per_second": 257.844,
      "eval_steps_per_second": 4.057,
      "step": 2331
    }
  ],
  "logging_steps": 500,
  "max_steps": 2331,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "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": 5.7235004676864e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}