trainer_state.json

{
  "best_metric": 0.8629333333333333,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0494/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": 139.5189971923828,
      "learning_rate": 9e-05,
      "loss": 2.8967,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.676,
      "eval_loss": 1.2021770477294922,
      "eval_runtime": 8.77,
      "eval_samples_per_second": 427.592,
      "eval_steps_per_second": 6.727,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 94.6420669555664,
      "learning_rate": 9e-05,
      "loss": 0.8127,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8032,
      "eval_loss": 0.6178784966468811,
      "eval_runtime": 7.8413,
      "eval_samples_per_second": 478.238,
      "eval_steps_per_second": 7.524,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 154.36166381835938,
      "learning_rate": 9e-05,
      "loss": 0.4779,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8349333333333333,
      "eval_loss": 0.5029668211936951,
      "eval_runtime": 8.2345,
      "eval_samples_per_second": 455.402,
      "eval_steps_per_second": 7.165,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 78.44804382324219,
      "learning_rate": 9e-05,
      "loss": 0.3188,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8421333333333333,
      "eval_loss": 0.4906027913093567,
      "eval_runtime": 8.5595,
      "eval_samples_per_second": 438.11,
      "eval_steps_per_second": 6.893,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 170.1049041748047,
      "learning_rate": 9e-05,
      "loss": 0.2218,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8576,
      "eval_loss": 0.4617575407028198,
      "eval_runtime": 7.919,
      "eval_samples_per_second": 473.547,
      "eval_steps_per_second": 7.45,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 146.71975708007812,
      "learning_rate": 9e-05,
      "loss": 0.1568,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8528,
      "eval_loss": 0.48444005846977234,
      "eval_runtime": 8.1381,
      "eval_samples_per_second": 460.795,
      "eval_steps_per_second": 7.25,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 109.54866790771484,
      "learning_rate": 9e-05,
      "loss": 0.1131,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8605333333333334,
      "eval_loss": 0.4938186705112457,
      "eval_runtime": 8.3286,
      "eval_samples_per_second": 450.256,
      "eval_steps_per_second": 7.084,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 135.7156524658203,
      "learning_rate": 9e-05,
      "loss": 0.0786,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8629333333333333,
      "eval_loss": 0.5153955817222595,
      "eval_runtime": 8.1652,
      "eval_samples_per_second": 459.266,
      "eval_steps_per_second": 7.226,
      "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
}