trainer_state.json

{
  "best_metric": 0.9112,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0254/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": 47.43699645996094,
      "learning_rate": 0.00047524221697560476,
      "loss": 1.2524,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8346666666666667,
      "eval_loss": 0.5589629411697388,
      "eval_runtime": 14.5731,
      "eval_samples_per_second": 257.323,
      "eval_steps_per_second": 4.049,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 27.448408126831055,
      "learning_rate": 0.0004058724504646834,
      "loss": 0.4027,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8688,
      "eval_loss": 0.4414268732070923,
      "eval_runtime": 14.535,
      "eval_samples_per_second": 257.999,
      "eval_steps_per_second": 4.059,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 21.33979606628418,
      "learning_rate": 0.0003056302334890786,
      "loss": 0.2218,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8810666666666667,
      "eval_loss": 0.4157167971134186,
      "eval_runtime": 14.6062,
      "eval_samples_per_second": 256.74,
      "eval_steps_per_second": 4.039,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 33.57265853881836,
      "learning_rate": 0.00019436976651092142,
      "loss": 0.1141,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9032,
      "eval_loss": 0.3615852892398834,
      "eval_runtime": 14.5489,
      "eval_samples_per_second": 257.752,
      "eval_steps_per_second": 4.055,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 22.222637176513672,
      "learning_rate": 9.412754953531663e-05,
      "loss": 0.0601,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9077333333333333,
      "eval_loss": 0.3519399166107178,
      "eval_runtime": 14.6542,
      "eval_samples_per_second": 255.899,
      "eval_steps_per_second": 4.026,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 30.614391326904297,
      "learning_rate": 2.4757783024395242e-05,
      "loss": 0.033,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9112,
      "eval_loss": 0.3562857508659363,
      "eval_runtime": 14.5614,
      "eval_samples_per_second": 257.53,
      "eval_steps_per_second": 4.052,
      "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
}