trainer_state.json

{
  "best_metric": 0.8786666666666667,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0168/checkpoints/checkpoint-2331",
  "epoch": 7.0,
  "eval_steps": 500,
  "global_step": 2331,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 70.74325561523438,
      "learning_rate": 7e-05,
      "loss": 3.1378,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7101333333333333,
      "eval_loss": 1.285015344619751,
      "eval_runtime": 13.2151,
      "eval_samples_per_second": 283.766,
      "eval_steps_per_second": 4.465,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 119.14641571044922,
      "learning_rate": 7e-05,
      "loss": 0.8326,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8221333333333334,
      "eval_loss": 0.5910533666610718,
      "eval_runtime": 12.2823,
      "eval_samples_per_second": 305.317,
      "eval_steps_per_second": 4.804,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 130.5901641845703,
      "learning_rate": 7e-05,
      "loss": 0.4794,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8586666666666667,
      "eval_loss": 0.4630683660507202,
      "eval_runtime": 12.2136,
      "eval_samples_per_second": 307.035,
      "eval_steps_per_second": 4.831,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 72.35430908203125,
      "learning_rate": 7e-05,
      "loss": 0.3275,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8658666666666667,
      "eval_loss": 0.43134889006614685,
      "eval_runtime": 12.285,
      "eval_samples_per_second": 305.251,
      "eval_steps_per_second": 4.803,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 78.9314193725586,
      "learning_rate": 7e-05,
      "loss": 0.2312,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8672,
      "eval_loss": 0.425525963306427,
      "eval_runtime": 12.2837,
      "eval_samples_per_second": 305.283,
      "eval_steps_per_second": 4.803,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 81.00181579589844,
      "learning_rate": 7e-05,
      "loss": 0.1593,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8757333333333334,
      "eval_loss": 0.4225448966026306,
      "eval_runtime": 12.1604,
      "eval_samples_per_second": 308.379,
      "eval_steps_per_second": 4.852,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 285.7117614746094,
      "learning_rate": 7e-05,
      "loss": 0.1087,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8786666666666667,
      "eval_loss": 0.4420524835586548,
      "eval_runtime": 12.1561,
      "eval_samples_per_second": 308.487,
      "eval_steps_per_second": 4.854,
      "step": 2331
    }
  ],
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 5.7235004676864e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}