trainer_state.json

{
  "best_metric": 0.896,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0406/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": 400.6116943359375,
      "learning_rate": 9e-05,
      "loss": 2.8124,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7536,
      "eval_loss": 0.9385801553726196,
      "eval_runtime": 12.5678,
      "eval_samples_per_second": 298.381,
      "eval_steps_per_second": 4.695,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 131.32725524902344,
      "learning_rate": 9e-05,
      "loss": 0.6784,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8509333333333333,
      "eval_loss": 0.5100927948951721,
      "eval_runtime": 12.5386,
      "eval_samples_per_second": 299.075,
      "eval_steps_per_second": 4.705,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 52.42090606689453,
      "learning_rate": 9e-05,
      "loss": 0.4102,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8754666666666666,
      "eval_loss": 0.40167927742004395,
      "eval_runtime": 12.7909,
      "eval_samples_per_second": 293.178,
      "eval_steps_per_second": 4.613,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 85.91487121582031,
      "learning_rate": 9e-05,
      "loss": 0.2878,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8885333333333333,
      "eval_loss": 0.36178287863731384,
      "eval_runtime": 12.6776,
      "eval_samples_per_second": 295.797,
      "eval_steps_per_second": 4.654,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 170.0952911376953,
      "learning_rate": 9e-05,
      "loss": 0.2108,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8917333333333334,
      "eval_loss": 0.34952616691589355,
      "eval_runtime": 12.6863,
      "eval_samples_per_second": 295.594,
      "eval_steps_per_second": 4.651,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 62.783241271972656,
      "learning_rate": 9e-05,
      "loss": 0.1596,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8954666666666666,
      "eval_loss": 0.3378992974758148,
      "eval_runtime": 12.7308,
      "eval_samples_per_second": 294.56,
      "eval_steps_per_second": 4.634,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 80.0831298828125,
      "learning_rate": 9e-05,
      "loss": 0.1144,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8952,
      "eval_loss": 0.36054569482803345,
      "eval_runtime": 12.8287,
      "eval_samples_per_second": 292.313,
      "eval_steps_per_second": 4.599,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 152.3560028076172,
      "learning_rate": 9e-05,
      "loss": 0.0968,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.896,
      "eval_loss": 0.3761001229286194,
      "eval_runtime": 12.6484,
      "eval_samples_per_second": 296.479,
      "eval_steps_per_second": 4.665,
      "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
}