trainer_state.json

{
  "best_metric": 0.876,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0651/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": 161.7594451904297,
      "learning_rate": 9.045084971874738e-05,
      "loss": 2.5502,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7856,
      "eval_loss": 0.8037369847297668,
      "eval_runtime": 7.2626,
      "eval_samples_per_second": 516.343,
      "eval_steps_per_second": 8.124,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 144.74325561523438,
      "learning_rate": 6.545084971874738e-05,
      "loss": 0.6284,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8549333333333333,
      "eval_loss": 0.48505595326423645,
      "eval_runtime": 7.2179,
      "eval_samples_per_second": 519.543,
      "eval_steps_per_second": 8.174,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 120.00006103515625,
      "learning_rate": 3.4549150281252636e-05,
      "loss": 0.4105,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8632,
      "eval_loss": 0.4462037682533264,
      "eval_runtime": 7.0846,
      "eval_samples_per_second": 529.318,
      "eval_steps_per_second": 8.328,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 212.01058959960938,
      "learning_rate": 9.549150281252633e-06,
      "loss": 0.3161,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8712,
      "eval_loss": 0.4113207757472992,
      "eval_runtime": 7.1162,
      "eval_samples_per_second": 526.967,
      "eval_steps_per_second": 8.291,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 52.600521087646484,
      "learning_rate": 0.0,
      "loss": 0.2799,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.876,
      "eval_loss": 0.407278448343277,
      "eval_runtime": 7.0414,
      "eval_samples_per_second": 532.567,
      "eval_steps_per_second": 8.379,
      "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
}