trainer_state.json

{
  "best_metric": 0.8744,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0155/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": 220.1686553955078,
      "learning_rate": 6.733578363789503e-05,
      "loss": 3.4266,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5757333333333333,
      "eval_loss": 1.981347918510437,
      "eval_runtime": 8.6416,
      "eval_samples_per_second": 433.945,
      "eval_steps_per_second": 6.827,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 203.3643035888672,
      "learning_rate": 5.974873734152915e-05,
      "loss": 1.1708,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7912,
      "eval_loss": 0.744491696357727,
      "eval_runtime": 8.6132,
      "eval_samples_per_second": 435.38,
      "eval_steps_per_second": 6.85,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 126.22608184814453,
      "learning_rate": 4.839392013277814e-05,
      "loss": 0.6292,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8274666666666667,
      "eval_loss": 0.5798434615135193,
      "eval_runtime": 8.5752,
      "eval_samples_per_second": 437.308,
      "eval_steps_per_second": 6.88,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 167.46487426757812,
      "learning_rate": 3.5e-05,
      "loss": 0.4626,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8624,
      "eval_loss": 0.46748799085617065,
      "eval_runtime": 9.0167,
      "eval_samples_per_second": 415.894,
      "eval_steps_per_second": 6.543,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 325.47515869140625,
      "learning_rate": 2.1606079867221858e-05,
      "loss": 0.3825,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8656,
      "eval_loss": 0.44140303134918213,
      "eval_runtime": 8.519,
      "eval_samples_per_second": 440.192,
      "eval_steps_per_second": 6.926,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 138.58892822265625,
      "learning_rate": 1.0251262658470837e-05,
      "loss": 0.3313,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8704,
      "eval_loss": 0.4366644024848938,
      "eval_runtime": 8.6689,
      "eval_samples_per_second": 432.581,
      "eval_steps_per_second": 6.806,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 69.9917984008789,
      "learning_rate": 2.664216362104964e-06,
      "loss": 0.3162,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8693333333333333,
      "eval_loss": 0.4417852759361267,
      "eval_runtime": 8.519,
      "eval_samples_per_second": 440.195,
      "eval_steps_per_second": 6.926,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 228.14263916015625,
      "learning_rate": 0.0,
      "loss": 0.3191,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8744,
      "eval_loss": 0.42603710293769836,
      "eval_runtime": 9.553,
      "eval_samples_per_second": 392.546,
      "eval_steps_per_second": 6.176,
      "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
}