trainer_state.json

{
  "best_metric": 0.8768,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0078/checkpoints/checkpoint-2664",
  "epoch": 8.0,
  "eval_steps": 500,
  "global_step": 2664,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 63.58607482910156,
      "learning_rate": 7e-05,
      "loss": 3.3706,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5845333333333333,
      "eval_loss": 1.9306870698928833,
      "eval_runtime": 15.1476,
      "eval_samples_per_second": 247.563,
      "eval_steps_per_second": 3.895,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 122.27293395996094,
      "learning_rate": 7e-05,
      "loss": 1.1319,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7658666666666667,
      "eval_loss": 0.792654812335968,
      "eval_runtime": 15.0255,
      "eval_samples_per_second": 249.576,
      "eval_steps_per_second": 3.927,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 78.60237884521484,
      "learning_rate": 7e-05,
      "loss": 0.6614,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8218666666666666,
      "eval_loss": 0.5740786194801331,
      "eval_runtime": 15.1626,
      "eval_samples_per_second": 247.32,
      "eval_steps_per_second": 3.891,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 111.03172302246094,
      "learning_rate": 7e-05,
      "loss": 0.4891,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8442666666666667,
      "eval_loss": 0.5021602511405945,
      "eval_runtime": 15.566,
      "eval_samples_per_second": 240.909,
      "eval_steps_per_second": 3.79,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 85.44111633300781,
      "learning_rate": 7e-05,
      "loss": 0.3938,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8570666666666666,
      "eval_loss": 0.4732164144515991,
      "eval_runtime": 15.1476,
      "eval_samples_per_second": 247.564,
      "eval_steps_per_second": 3.895,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 53.554115295410156,
      "learning_rate": 7e-05,
      "loss": 0.3211,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8621333333333333,
      "eval_loss": 0.43745866417884827,
      "eval_runtime": 15.0008,
      "eval_samples_per_second": 249.986,
      "eval_steps_per_second": 3.933,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 81.97529602050781,
      "learning_rate": 7e-05,
      "loss": 0.2558,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8701333333333333,
      "eval_loss": 0.4195326864719391,
      "eval_runtime": 15.0051,
      "eval_samples_per_second": 249.915,
      "eval_steps_per_second": 3.932,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 66.70734405517578,
      "learning_rate": 7e-05,
      "loss": 0.2046,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8768,
      "eval_loss": 0.41398924589157104,
      "eval_runtime": 15.0891,
      "eval_samples_per_second": 248.524,
      "eval_steps_per_second": 3.91,
      "step": 2664
    }
  ],
  "logging_steps": 500,
  "max_steps": 2664,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "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": 6.5411433916416e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}