trainer_state.json

{
  "best_metric": 0.8826666666666667,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0857/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": 97.82379150390625,
      "learning_rate": 9.619397662556435e-05,
      "loss": 2.8464,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7322666666666666,
      "eval_loss": 1.0170294046401978,
      "eval_runtime": 14.586,
      "eval_samples_per_second": 257.096,
      "eval_steps_per_second": 4.045,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 133.21177673339844,
      "learning_rate": 8.535533905932738e-05,
      "loss": 0.7715,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8290666666666666,
      "eval_loss": 0.5703845024108887,
      "eval_runtime": 14.2959,
      "eval_samples_per_second": 262.313,
      "eval_steps_per_second": 4.127,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 104.28449249267578,
      "learning_rate": 6.91341716182545e-05,
      "loss": 0.5064,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8589333333333333,
      "eval_loss": 0.4567476511001587,
      "eval_runtime": 14.3751,
      "eval_samples_per_second": 260.868,
      "eval_steps_per_second": 4.104,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 69.7591781616211,
      "learning_rate": 5e-05,
      "loss": 0.3938,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8741333333333333,
      "eval_loss": 0.4031757116317749,
      "eval_runtime": 14.2596,
      "eval_samples_per_second": 262.98,
      "eval_steps_per_second": 4.138,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 144.24362182617188,
      "learning_rate": 3.086582838174551e-05,
      "loss": 0.3122,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8778666666666667,
      "eval_loss": 0.37770023941993713,
      "eval_runtime": 14.1629,
      "eval_samples_per_second": 264.777,
      "eval_steps_per_second": 4.166,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 74.13427734375,
      "learning_rate": 1.4644660940672627e-05,
      "loss": 0.2795,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8770666666666667,
      "eval_loss": 0.3872314393520355,
      "eval_runtime": 14.2841,
      "eval_samples_per_second": 262.53,
      "eval_steps_per_second": 4.13,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 56.14397048950195,
      "learning_rate": 3.8060233744356633e-06,
      "loss": 0.2551,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8765333333333334,
      "eval_loss": 0.3905186951160431,
      "eval_runtime": 14.208,
      "eval_samples_per_second": 263.935,
      "eval_steps_per_second": 4.153,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 82.00831604003906,
      "learning_rate": 0.0,
      "loss": 0.2417,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8826666666666667,
      "eval_loss": 0.37602078914642334,
      "eval_runtime": 14.2639,
      "eval_samples_per_second": 262.902,
      "eval_steps_per_second": 4.136,
      "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
}