trainer_state.json

{
  "best_metric": 0.8354666666666667,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0839/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": 61.488101959228516,
      "learning_rate": 2.9095389311788626e-05,
      "loss": 3.7889,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5384,
      "eval_loss": 3.517068862915039,
      "eval_runtime": 8.9544,
      "eval_samples_per_second": 418.791,
      "eval_steps_per_second": 6.589,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 130.88818359375,
      "learning_rate": 2.649066664678467e-05,
      "loss": 2.8587,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6746666666666666,
      "eval_loss": 2.2239584922790527,
      "eval_runtime": 7.5708,
      "eval_samples_per_second": 495.321,
      "eval_steps_per_second": 7.793,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 128.44947814941406,
      "learning_rate": 2.25e-05,
      "loss": 1.4432,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7677333333333334,
      "eval_loss": 0.9581058025360107,
      "eval_runtime": 7.5777,
      "eval_samples_per_second": 494.87,
      "eval_steps_per_second": 7.786,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 125.2374038696289,
      "learning_rate": 1.760472266500396e-05,
      "loss": 0.7976,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8085333333333333,
      "eval_loss": 0.6913534998893738,
      "eval_runtime": 7.5353,
      "eval_samples_per_second": 497.66,
      "eval_steps_per_second": 7.83,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 146.4196014404297,
      "learning_rate": 1.2395277334996045e-05,
      "loss": 0.6176,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8237333333333333,
      "eval_loss": 0.6058608293533325,
      "eval_runtime": 7.5321,
      "eval_samples_per_second": 497.869,
      "eval_steps_per_second": 7.833,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 266.3782043457031,
      "learning_rate": 7.500000000000004e-06,
      "loss": 0.5303,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8234666666666667,
      "eval_loss": 0.5956773161888123,
      "eval_runtime": 7.5545,
      "eval_samples_per_second": 496.394,
      "eval_steps_per_second": 7.81,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 162.99118041992188,
      "learning_rate": 3.5093333532153316e-06,
      "loss": 0.4868,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8333333333333334,
      "eval_loss": 0.5470821857452393,
      "eval_runtime": 7.8149,
      "eval_samples_per_second": 479.855,
      "eval_steps_per_second": 7.55,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 274.5069885253906,
      "learning_rate": 9.046106882113753e-07,
      "loss": 0.4619,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8354666666666667,
      "eval_loss": 0.5387735962867737,
      "eval_runtime": 7.7756,
      "eval_samples_per_second": 482.28,
      "eval_steps_per_second": 7.588,
      "step": 2664
    }
  ],
  "logging_steps": 500,
  "max_steps": 2997,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 6.5411433916416e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}