trainer_state.json

{
  "best_metric": 0.8744,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0280/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": 30.431625366210938,
      "learning_rate": 0.0005,
      "loss": 1.3135,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8186666666666667,
      "eval_loss": 0.5980513095855713,
      "eval_runtime": 7.9649,
      "eval_samples_per_second": 470.818,
      "eval_steps_per_second": 7.408,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 34.672637939453125,
      "learning_rate": 0.0005,
      "loss": 0.4723,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8482666666666666,
      "eval_loss": 0.49764516949653625,
      "eval_runtime": 7.0971,
      "eval_samples_per_second": 528.388,
      "eval_steps_per_second": 8.313,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 24.00688362121582,
      "learning_rate": 0.0005,
      "loss": 0.3152,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8450666666666666,
      "eval_loss": 0.5045941472053528,
      "eval_runtime": 7.5374,
      "eval_samples_per_second": 497.521,
      "eval_steps_per_second": 7.828,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 26.704042434692383,
      "learning_rate": 0.0005,
      "loss": 0.2269,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8613333333333333,
      "eval_loss": 0.48389551043510437,
      "eval_runtime": 7.1202,
      "eval_samples_per_second": 526.673,
      "eval_steps_per_second": 8.286,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 26.29146385192871,
      "learning_rate": 0.0005,
      "loss": 0.172,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8602666666666666,
      "eval_loss": 0.5297295451164246,
      "eval_runtime": 7.1311,
      "eval_samples_per_second": 525.862,
      "eval_steps_per_second": 8.274,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 23.196945190429688,
      "learning_rate": 0.0005,
      "loss": 0.1241,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8744,
      "eval_loss": 0.4917345643043518,
      "eval_runtime": 7.5849,
      "eval_samples_per_second": 494.401,
      "eval_steps_per_second": 7.779,
      "step": 1998
    }
  ],
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 4.9058575437312e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}