trainer_state.json

{
  "best_metric": 0.8717333333333334,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0649/checkpoints/checkpoint-1665",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 1665,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 79.84327697753906,
      "learning_rate": 5.6e-05,
      "loss": 3.3392,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6618666666666667,
      "eval_loss": 2.021726131439209,
      "eval_runtime": 9.6239,
      "eval_samples_per_second": 389.657,
      "eval_steps_per_second": 6.131,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 63.79200744628906,
      "learning_rate": 4.2e-05,
      "loss": 1.0581,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8373333333333334,
      "eval_loss": 0.6312043070793152,
      "eval_runtime": 9.1493,
      "eval_samples_per_second": 409.866,
      "eval_steps_per_second": 6.449,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 63.71295928955078,
      "learning_rate": 2.8e-05,
      "loss": 0.5658,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8584,
      "eval_loss": 0.500495970249176,
      "eval_runtime": 9.0639,
      "eval_samples_per_second": 413.73,
      "eval_steps_per_second": 6.509,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 222.4241485595703,
      "learning_rate": 1.4e-05,
      "loss": 0.4358,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8690666666666667,
      "eval_loss": 0.444312185049057,
      "eval_runtime": 8.9802,
      "eval_samples_per_second": 417.587,
      "eval_steps_per_second": 6.57,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 154.5330352783203,
      "learning_rate": 0.0,
      "loss": 0.3845,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8717333333333334,
      "eval_loss": 0.4392085373401642,
      "eval_runtime": 8.8742,
      "eval_samples_per_second": 422.575,
      "eval_steps_per_second": 6.649,
      "step": 1665
    }
  ],
  "logging_steps": 500,
  "max_steps": 1665,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "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": 4.088214619776e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}