trainer_state.json

{
  "best_metric": 0.8677333333333334,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0754/checkpoints/checkpoint-1665",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 1665,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 129.32032775878906,
      "learning_rate": 6.331559480312315e-05,
      "loss": 3.2784,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6698666666666667,
      "eval_loss": 1.764281988143921,
      "eval_runtime": 7.906,
      "eval_samples_per_second": 474.322,
      "eval_steps_per_second": 7.463,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 121.84637451171875,
      "learning_rate": 4.581559480312316e-05,
      "loss": 0.913,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8309333333333333,
      "eval_loss": 0.5979891419410706,
      "eval_runtime": 7.6645,
      "eval_samples_per_second": 489.267,
      "eval_steps_per_second": 7.698,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 233.96388244628906,
      "learning_rate": 2.4184405196876842e-05,
      "loss": 0.5003,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8565333333333334,
      "eval_loss": 0.45693859457969666,
      "eval_runtime": 7.8667,
      "eval_samples_per_second": 476.691,
      "eval_steps_per_second": 7.5,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 507.80218505859375,
      "learning_rate": 6.684405196876842e-06,
      "loss": 0.4028,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8648,
      "eval_loss": 0.43168923258781433,
      "eval_runtime": 7.6914,
      "eval_samples_per_second": 487.556,
      "eval_steps_per_second": 7.671,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 76.29035186767578,
      "learning_rate": 0.0,
      "loss": 0.3704,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8677333333333334,
      "eval_loss": 0.41839250922203064,
      "eval_runtime": 8.1836,
      "eval_samples_per_second": 458.232,
      "eval_steps_per_second": 7.21,
      "step": 1665
    }
  ],
  "logging_steps": 500,
  "max_steps": 1665,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "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": 4.088214619776e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}