trainer_state.json

{
  "best_metric": 0.8976,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0036/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": 112.27693939208984,
      "learning_rate": 9e-05,
      "loss": 2.8614,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7469333333333333,
      "eval_loss": 1.0130677223205566,
      "eval_runtime": 15.2303,
      "eval_samples_per_second": 246.22,
      "eval_steps_per_second": 3.874,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 67.82659912109375,
      "learning_rate": 9e-05,
      "loss": 0.7227,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8408,
      "eval_loss": 0.5617716312408447,
      "eval_runtime": 15.2539,
      "eval_samples_per_second": 245.839,
      "eval_steps_per_second": 3.868,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 87.87979888916016,
      "learning_rate": 9e-05,
      "loss": 0.4686,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8632,
      "eval_loss": 0.46093279123306274,
      "eval_runtime": 15.156,
      "eval_samples_per_second": 247.426,
      "eval_steps_per_second": 3.893,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 68.3366470336914,
      "learning_rate": 9e-05,
      "loss": 0.3449,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8757333333333334,
      "eval_loss": 0.4452570080757141,
      "eval_runtime": 15.436,
      "eval_samples_per_second": 242.938,
      "eval_steps_per_second": 3.822,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 67.59040832519531,
      "learning_rate": 9e-05,
      "loss": 0.2611,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8778666666666667,
      "eval_loss": 0.4204079508781433,
      "eval_runtime": 15.2169,
      "eval_samples_per_second": 246.437,
      "eval_steps_per_second": 3.877,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 177.40322875976562,
      "learning_rate": 9e-05,
      "loss": 0.2037,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8890666666666667,
      "eval_loss": 0.39576661586761475,
      "eval_runtime": 15.0206,
      "eval_samples_per_second": 249.657,
      "eval_steps_per_second": 3.928,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 71.08078002929688,
      "learning_rate": 9e-05,
      "loss": 0.1609,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8909333333333334,
      "eval_loss": 0.42112642526626587,
      "eval_runtime": 15.1176,
      "eval_samples_per_second": 248.055,
      "eval_steps_per_second": 3.903,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 72.15408325195312,
      "learning_rate": 9e-05,
      "loss": 0.1254,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8925333333333333,
      "eval_loss": 0.4110449254512787,
      "eval_runtime": 16.04,
      "eval_samples_per_second": 233.79,
      "eval_steps_per_second": 3.678,
      "step": 2664
    },
    {
      "epoch": 9.0,
      "grad_norm": 237.10768127441406,
      "learning_rate": 9e-05,
      "loss": 0.1019,
      "step": 2997
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.8976,
      "eval_loss": 0.4198263883590698,
      "eval_runtime": 15.2,
      "eval_samples_per_second": 246.711,
      "eval_steps_per_second": 3.882,
      "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
}