trainer_state.json

{
  "best_metric": 0.8834666666666666,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0145/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": 77.0848617553711,
      "learning_rate": 4.849231551964771e-05,
      "loss": 3.6583,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5592,
      "eval_loss": 2.9445767402648926,
      "eval_runtime": 8.2146,
      "eval_samples_per_second": 456.505,
      "eval_steps_per_second": 7.182,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 197.98876953125,
      "learning_rate": 4.415111107797445e-05,
      "loss": 1.5345,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7957333333333333,
      "eval_loss": 0.7878127098083496,
      "eval_runtime": 8.1285,
      "eval_samples_per_second": 461.341,
      "eval_steps_per_second": 7.258,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 125.95204162597656,
      "learning_rate": 3.7500000000000003e-05,
      "loss": 0.6718,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.852,
      "eval_loss": 0.5368257164955139,
      "eval_runtime": 7.9503,
      "eval_samples_per_second": 471.68,
      "eval_steps_per_second": 7.421,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 164.25343322753906,
      "learning_rate": 2.9341204441673266e-05,
      "loss": 0.4859,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8682666666666666,
      "eval_loss": 0.4547681212425232,
      "eval_runtime": 8.6778,
      "eval_samples_per_second": 432.136,
      "eval_steps_per_second": 6.799,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 119.71442413330078,
      "learning_rate": 2.0658795558326743e-05,
      "loss": 0.3884,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8701333333333333,
      "eval_loss": 0.42370307445526123,
      "eval_runtime": 8.2321,
      "eval_samples_per_second": 455.535,
      "eval_steps_per_second": 7.167,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 233.80734252929688,
      "learning_rate": 1.2500000000000006e-05,
      "loss": 0.3406,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8786666666666667,
      "eval_loss": 0.40217188000679016,
      "eval_runtime": 8.1734,
      "eval_samples_per_second": 458.804,
      "eval_steps_per_second": 7.219,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 377.4540710449219,
      "learning_rate": 5.848888922025553e-06,
      "loss": 0.3045,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8805333333333333,
      "eval_loss": 0.3989948630332947,
      "eval_runtime": 8.1726,
      "eval_samples_per_second": 458.85,
      "eval_steps_per_second": 7.219,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 57.223976135253906,
      "learning_rate": 1.5076844803522922e-06,
      "loss": 0.2893,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8834666666666666,
      "eval_loss": 0.38946855068206787,
      "eval_runtime": 8.5579,
      "eval_samples_per_second": 438.191,
      "eval_steps_per_second": 6.894,
      "step": 2664
    }
  ],
  "logging_steps": 500,
  "max_steps": 2997,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9,
  "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": 6.5411433916416e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}