trainer_state.json

{
  "best_metric": 0.8349333333333333,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0122/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": 120.99430847167969,
      "learning_rate": 2.8858192987669303e-05,
      "loss": 3.7946,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.4792,
      "eval_loss": 3.533592700958252,
      "eval_runtime": 12.8248,
      "eval_samples_per_second": 292.402,
      "eval_steps_per_second": 4.6,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 199.9960174560547,
      "learning_rate": 2.5606601717798212e-05,
      "loss": 2.9212,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6194666666666667,
      "eval_loss": 2.212885618209839,
      "eval_runtime": 12.7139,
      "eval_samples_per_second": 294.953,
      "eval_steps_per_second": 4.641,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 160.40228271484375,
      "learning_rate": 2.074025148547635e-05,
      "loss": 1.5101,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7605333333333333,
      "eval_loss": 1.0271196365356445,
      "eval_runtime": 12.6638,
      "eval_samples_per_second": 296.12,
      "eval_steps_per_second": 4.659,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 466.71343994140625,
      "learning_rate": 1.5e-05,
      "loss": 0.909,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8098666666666666,
      "eval_loss": 0.7619207501411438,
      "eval_runtime": 12.6091,
      "eval_samples_per_second": 297.404,
      "eval_steps_per_second": 4.679,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 340.89923095703125,
      "learning_rate": 9.259748514523654e-06,
      "loss": 0.7227,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8229333333333333,
      "eval_loss": 0.6568772196769714,
      "eval_runtime": 12.7593,
      "eval_samples_per_second": 293.903,
      "eval_steps_per_second": 4.624,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 148.53634643554688,
      "learning_rate": 4.393398282201788e-06,
      "loss": 0.6351,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8298666666666666,
      "eval_loss": 0.6373768448829651,
      "eval_runtime": 12.573,
      "eval_samples_per_second": 298.258,
      "eval_steps_per_second": 4.693,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 95.5239028930664,
      "learning_rate": 1.1418070123306989e-06,
      "loss": 0.6001,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8317333333333333,
      "eval_loss": 0.6264415383338928,
      "eval_runtime": 12.5634,
      "eval_samples_per_second": 298.485,
      "eval_steps_per_second": 4.696,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 142.74075317382812,
      "learning_rate": 0.0,
      "loss": 0.5957,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8349333333333333,
      "eval_loss": 0.6046580672264099,
      "eval_runtime": 12.7179,
      "eval_samples_per_second": 294.86,
      "eval_steps_per_second": 4.639,
      "step": 2664
    }
  ],
  "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": true
      },
      "attributes": {}
    }
  },
  "total_flos": 6.5411433916416e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}