trainer_state.json

{
  "best_metric": 0.8997333333333334,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0973/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": 73.84014892578125,
      "learning_rate": 5e-05,
      "loss": 3.5766,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6088,
      "eval_loss": 2.7272493839263916,
      "eval_runtime": 14.5774,
      "eval_samples_per_second": 257.248,
      "eval_steps_per_second": 4.047,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 140.62994384765625,
      "learning_rate": 5e-05,
      "loss": 1.3841,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8136,
      "eval_loss": 0.688335657119751,
      "eval_runtime": 14.4771,
      "eval_samples_per_second": 259.03,
      "eval_steps_per_second": 4.075,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 92.28995513916016,
      "learning_rate": 5e-05,
      "loss": 0.5821,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8512,
      "eval_loss": 0.48877739906311035,
      "eval_runtime": 13.8101,
      "eval_samples_per_second": 271.541,
      "eval_steps_per_second": 4.272,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 49.408164978027344,
      "learning_rate": 5e-05,
      "loss": 0.4044,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8701333333333333,
      "eval_loss": 0.42334824800491333,
      "eval_runtime": 13.5593,
      "eval_samples_per_second": 276.563,
      "eval_steps_per_second": 4.351,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 42.78351593017578,
      "learning_rate": 5e-05,
      "loss": 0.3132,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8829333333333333,
      "eval_loss": 0.39217475056648254,
      "eval_runtime": 13.4851,
      "eval_samples_per_second": 278.084,
      "eval_steps_per_second": 4.375,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 147.5037841796875,
      "learning_rate": 5e-05,
      "loss": 0.2451,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8906666666666667,
      "eval_loss": 0.3477572798728943,
      "eval_runtime": 13.6778,
      "eval_samples_per_second": 274.167,
      "eval_steps_per_second": 4.314,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 216.37669372558594,
      "learning_rate": 5e-05,
      "loss": 0.1954,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8952,
      "eval_loss": 0.34507444500923157,
      "eval_runtime": 13.4682,
      "eval_samples_per_second": 278.434,
      "eval_steps_per_second": 4.381,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 96.99578857421875,
      "learning_rate": 5e-05,
      "loss": 0.1601,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8997333333333334,
      "eval_loss": 0.35078391432762146,
      "eval_runtime": 13.4673,
      "eval_samples_per_second": 278.453,
      "eval_steps_per_second": 4.381,
      "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
}