trainer_state.json

{
  "best_metric": 0.8872,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0913/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": 170.23446655273438,
      "learning_rate": 7e-05,
      "loss": 3.39,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.572,
      "eval_loss": 2.019514560699463,
      "eval_runtime": 8.3159,
      "eval_samples_per_second": 450.943,
      "eval_steps_per_second": 7.095,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 158.74302673339844,
      "learning_rate": 7e-05,
      "loss": 1.148,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7856,
      "eval_loss": 0.7316609621047974,
      "eval_runtime": 8.408,
      "eval_samples_per_second": 446.002,
      "eval_steps_per_second": 7.017,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 84.04093170166016,
      "learning_rate": 7e-05,
      "loss": 0.6278,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8362666666666667,
      "eval_loss": 0.5347887277603149,
      "eval_runtime": 8.3568,
      "eval_samples_per_second": 448.735,
      "eval_steps_per_second": 7.06,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 83.478759765625,
      "learning_rate": 7e-05,
      "loss": 0.4687,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8704,
      "eval_loss": 0.4368656277656555,
      "eval_runtime": 8.3683,
      "eval_samples_per_second": 448.118,
      "eval_steps_per_second": 7.05,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 125.67264556884766,
      "learning_rate": 7e-05,
      "loss": 0.3677,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8762666666666666,
      "eval_loss": 0.41281795501708984,
      "eval_runtime": 8.525,
      "eval_samples_per_second": 439.884,
      "eval_steps_per_second": 6.921,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 165.23553466796875,
      "learning_rate": 7e-05,
      "loss": 0.3057,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8693333333333333,
      "eval_loss": 0.41758665442466736,
      "eval_runtime": 8.2939,
      "eval_samples_per_second": 452.14,
      "eval_steps_per_second": 7.114,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 78.7149887084961,
      "learning_rate": 7e-05,
      "loss": 0.2485,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8786666666666667,
      "eval_loss": 0.4075377583503723,
      "eval_runtime": 8.378,
      "eval_samples_per_second": 447.601,
      "eval_steps_per_second": 7.042,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 47.93105697631836,
      "learning_rate": 7e-05,
      "loss": 0.2128,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8872,
      "eval_loss": 0.3883171081542969,
      "eval_runtime": 8.1553,
      "eval_samples_per_second": 459.825,
      "eval_steps_per_second": 7.235,
      "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
}