trainer_state.json

{
  "best_metric": 0.9045333333333333,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0326/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": 110.76824188232422,
      "learning_rate": 7e-05,
      "loss": 3.2489,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6906666666666667,
      "eval_loss": 1.5407238006591797,
      "eval_runtime": 12.8612,
      "eval_samples_per_second": 291.574,
      "eval_steps_per_second": 4.587,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 100.1020736694336,
      "learning_rate": 7e-05,
      "loss": 0.88,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8266666666666667,
      "eval_loss": 0.5969424247741699,
      "eval_runtime": 12.8835,
      "eval_samples_per_second": 291.07,
      "eval_steps_per_second": 4.579,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 132.7123565673828,
      "learning_rate": 7e-05,
      "loss": 0.4768,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8658666666666667,
      "eval_loss": 0.43194231390953064,
      "eval_runtime": 12.8731,
      "eval_samples_per_second": 291.306,
      "eval_steps_per_second": 4.583,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 78.180908203125,
      "learning_rate": 7e-05,
      "loss": 0.3299,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8773333333333333,
      "eval_loss": 0.4057547152042389,
      "eval_runtime": 13.4707,
      "eval_samples_per_second": 278.381,
      "eval_steps_per_second": 4.38,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 157.09815979003906,
      "learning_rate": 7e-05,
      "loss": 0.2495,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8810666666666667,
      "eval_loss": 0.39550158381462097,
      "eval_runtime": 12.9907,
      "eval_samples_per_second": 288.668,
      "eval_steps_per_second": 4.542,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 142.90969848632812,
      "learning_rate": 7e-05,
      "loss": 0.1885,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8896,
      "eval_loss": 0.36165547370910645,
      "eval_runtime": 12.8351,
      "eval_samples_per_second": 292.167,
      "eval_steps_per_second": 4.597,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 123.25326538085938,
      "learning_rate": 7e-05,
      "loss": 0.1418,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8930666666666667,
      "eval_loss": 0.35009920597076416,
      "eval_runtime": 12.8159,
      "eval_samples_per_second": 292.605,
      "eval_steps_per_second": 4.604,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 207.27035522460938,
      "learning_rate": 7e-05,
      "loss": 0.1088,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9045333333333333,
      "eval_loss": 0.3536478877067566,
      "eval_runtime": 12.7392,
      "eval_samples_per_second": 294.367,
      "eval_steps_per_second": 4.631,
      "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
}