trainer_state.json

{
  "best_metric": 0.8997333333333334,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0784/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": 173.60140991210938,
      "learning_rate": 9e-05,
      "loss": 2.8011,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7597333333333334,
      "eval_loss": 0.958266019821167,
      "eval_runtime": 14.8588,
      "eval_samples_per_second": 252.376,
      "eval_steps_per_second": 3.971,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 81.95524597167969,
      "learning_rate": 9e-05,
      "loss": 0.6928,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8472,
      "eval_loss": 0.5162932276725769,
      "eval_runtime": 14.4542,
      "eval_samples_per_second": 259.44,
      "eval_steps_per_second": 4.082,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 100.34757995605469,
      "learning_rate": 9e-05,
      "loss": 0.4079,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8728,
      "eval_loss": 0.41804009675979614,
      "eval_runtime": 14.327,
      "eval_samples_per_second": 261.744,
      "eval_steps_per_second": 4.118,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 67.30180358886719,
      "learning_rate": 9e-05,
      "loss": 0.2812,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8952,
      "eval_loss": 0.3425828814506531,
      "eval_runtime": 14.6414,
      "eval_samples_per_second": 256.124,
      "eval_steps_per_second": 4.03,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 72.84265899658203,
      "learning_rate": 9e-05,
      "loss": 0.2013,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8997333333333334,
      "eval_loss": 0.3541368544101715,
      "eval_runtime": 14.4012,
      "eval_samples_per_second": 260.394,
      "eval_steps_per_second": 4.097,
      "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
}