trainer_state.json

{
  "best_metric": 0.8936,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0680/checkpoints/checkpoint-1998",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 1998,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 106.45153045654297,
      "learning_rate": 8.554359905560886e-05,
      "loss": 2.9978,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7290666666666666,
      "eval_loss": 1.1854989528656006,
      "eval_runtime": 9.4272,
      "eval_samples_per_second": 397.787,
      "eval_steps_per_second": 6.259,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 100.78570556640625,
      "learning_rate": 7.305704108364301e-05,
      "loss": 0.816,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8413333333333334,
      "eval_loss": 0.5490705370903015,
      "eval_runtime": 9.4936,
      "eval_samples_per_second": 395.002,
      "eval_steps_per_second": 6.215,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 154.3232421875,
      "learning_rate": 5.501344202803415e-05,
      "loss": 0.5123,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8701333333333333,
      "eval_loss": 0.4455411434173584,
      "eval_runtime": 10.4286,
      "eval_samples_per_second": 359.587,
      "eval_steps_per_second": 5.658,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 69.76869201660156,
      "learning_rate": 3.498655797196586e-05,
      "loss": 0.4093,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8816,
      "eval_loss": 0.40363985300064087,
      "eval_runtime": 9.5478,
      "eval_samples_per_second": 392.76,
      "eval_steps_per_second": 6.179,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 133.90968322753906,
      "learning_rate": 1.6942958916356995e-05,
      "loss": 0.3452,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8885333333333333,
      "eval_loss": 0.38003969192504883,
      "eval_runtime": 9.5776,
      "eval_samples_per_second": 391.537,
      "eval_steps_per_second": 6.16,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 80.4434585571289,
      "learning_rate": 4.456400944391144e-06,
      "loss": 0.306,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8936,
      "eval_loss": 0.3761504292488098,
      "eval_runtime": 9.5514,
      "eval_samples_per_second": 392.614,
      "eval_steps_per_second": 6.177,
      "step": 1998
    }
  ],
  "logging_steps": 500,
  "max_steps": 2331,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "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": 4.9058575437312e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}