trainer_state.json

{
  "best_metric": 0.8746666666666667,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0108/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": 60.78998565673828,
      "learning_rate": 6.331559480312315e-05,
      "loss": 3.255,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6810666666666667,
      "eval_loss": 1.677929401397705,
      "eval_runtime": 14.1581,
      "eval_samples_per_second": 264.865,
      "eval_steps_per_second": 4.167,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 136.15011596679688,
      "learning_rate": 4.581559480312316e-05,
      "loss": 0.9284,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8448,
      "eval_loss": 0.6012980341911316,
      "eval_runtime": 14.1215,
      "eval_samples_per_second": 265.552,
      "eval_steps_per_second": 4.178,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 527.3392944335938,
      "learning_rate": 2.4184405196876842e-05,
      "loss": 0.4923,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8666666666666667,
      "eval_loss": 0.47995060682296753,
      "eval_runtime": 14.1752,
      "eval_samples_per_second": 264.547,
      "eval_steps_per_second": 4.162,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 183.74246215820312,
      "learning_rate": 6.684405196876842e-06,
      "loss": 0.3702,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8736,
      "eval_loss": 0.44717735052108765,
      "eval_runtime": 14.1425,
      "eval_samples_per_second": 265.158,
      "eval_steps_per_second": 4.172,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 107.67285919189453,
      "learning_rate": 0.0,
      "loss": 0.321,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8746666666666667,
      "eval_loss": 0.44898951053619385,
      "eval_runtime": 14.7967,
      "eval_samples_per_second": 253.434,
      "eval_steps_per_second": 3.987,
      "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
}