trainer_state.json

{
  "best_metric": 0.876,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0071/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": 97.49886322021484,
      "learning_rate": 5e-05,
      "loss": 3.5792,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5106666666666667,
      "eval_loss": 2.724250555038452,
      "eval_runtime": 12.6724,
      "eval_samples_per_second": 295.918,
      "eval_steps_per_second": 4.656,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 73.82286071777344,
      "learning_rate": 5e-05,
      "loss": 1.5178,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7701333333333333,
      "eval_loss": 0.8667422533035278,
      "eval_runtime": 12.4628,
      "eval_samples_per_second": 300.896,
      "eval_steps_per_second": 4.734,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 131.2230224609375,
      "learning_rate": 5e-05,
      "loss": 0.7191,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8122666666666667,
      "eval_loss": 0.606163740158081,
      "eval_runtime": 12.5026,
      "eval_samples_per_second": 299.938,
      "eval_steps_per_second": 4.719,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 213.77517700195312,
      "learning_rate": 5e-05,
      "loss": 0.5098,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.848,
      "eval_loss": 0.4969116449356079,
      "eval_runtime": 12.4596,
      "eval_samples_per_second": 300.972,
      "eval_steps_per_second": 4.735,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 79.2835693359375,
      "learning_rate": 5e-05,
      "loss": 0.3944,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8613333333333333,
      "eval_loss": 0.45226866006851196,
      "eval_runtime": 12.4813,
      "eval_samples_per_second": 300.451,
      "eval_steps_per_second": 4.727,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 176.5719757080078,
      "learning_rate": 5e-05,
      "loss": 0.3183,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8661333333333333,
      "eval_loss": 0.4227029085159302,
      "eval_runtime": 12.7277,
      "eval_samples_per_second": 294.632,
      "eval_steps_per_second": 4.636,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 168.22218322753906,
      "learning_rate": 5e-05,
      "loss": 0.2498,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8733333333333333,
      "eval_loss": 0.406497061252594,
      "eval_runtime": 12.4254,
      "eval_samples_per_second": 301.802,
      "eval_steps_per_second": 4.748,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 119.15180206298828,
      "learning_rate": 5e-05,
      "loss": 0.2108,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.876,
      "eval_loss": 0.4012466371059418,
      "eval_runtime": 12.4913,
      "eval_samples_per_second": 300.208,
      "eval_steps_per_second": 4.723,
      "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
}