trainer_state.json

{
  "best_metric": 0.8992,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0506/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": 43.333251953125,
      "learning_rate": 0.0002713525491562421,
      "loss": 1.5586,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8397333333333333,
      "eval_loss": 0.5417495965957642,
      "eval_runtime": 5.0299,
      "eval_samples_per_second": 745.539,
      "eval_steps_per_second": 11.73,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 30.934419631958008,
      "learning_rate": 0.0001963525491562421,
      "loss": 0.4,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8765333333333334,
      "eval_loss": 0.4059755802154541,
      "eval_runtime": 4.9757,
      "eval_samples_per_second": 753.658,
      "eval_steps_per_second": 11.858,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 69.62355041503906,
      "learning_rate": 0.0001036474508437579,
      "loss": 0.1852,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8898666666666667,
      "eval_loss": 0.38331538438796997,
      "eval_runtime": 4.9511,
      "eval_samples_per_second": 757.405,
      "eval_steps_per_second": 11.917,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 34.97834014892578,
      "learning_rate": 2.8647450843757897e-05,
      "loss": 0.093,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8941333333333333,
      "eval_loss": 0.36487674713134766,
      "eval_runtime": 4.9495,
      "eval_samples_per_second": 757.651,
      "eval_steps_per_second": 11.92,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 35.121368408203125,
      "learning_rate": 0.0,
      "loss": 0.052,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8992,
      "eval_loss": 0.35113972425460815,
      "eval_runtime": 5.0155,
      "eval_samples_per_second": 747.676,
      "eval_steps_per_second": 11.763,
      "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
}