trainer_state.json

{
  "best_metric": 0.8728,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0924/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": 106.87415313720703,
      "learning_rate": 8.657457896300791e-05,
      "loss": 3.0753,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6757333333333333,
      "eval_loss": 1.3192486763000488,
      "eval_runtime": 13.2514,
      "eval_samples_per_second": 282.989,
      "eval_steps_per_second": 4.452,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 173.34567260742188,
      "learning_rate": 7.681980515339464e-05,
      "loss": 0.8873,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.812,
      "eval_loss": 0.6386834383010864,
      "eval_runtime": 13.2131,
      "eval_samples_per_second": 283.808,
      "eval_steps_per_second": 4.465,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 62.949188232421875,
      "learning_rate": 6.222075445642905e-05,
      "loss": 0.5665,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8474666666666667,
      "eval_loss": 0.499980092048645,
      "eval_runtime": 14.2812,
      "eval_samples_per_second": 262.583,
      "eval_steps_per_second": 4.131,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 133.9521484375,
      "learning_rate": 4.5e-05,
      "loss": 0.4407,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8592,
      "eval_loss": 0.4622727930545807,
      "eval_runtime": 13.3729,
      "eval_samples_per_second": 280.417,
      "eval_steps_per_second": 4.412,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 125.10379791259766,
      "learning_rate": 2.7779245543570963e-05,
      "loss": 0.3707,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8650666666666667,
      "eval_loss": 0.44721513986587524,
      "eval_runtime": 13.1949,
      "eval_samples_per_second": 284.2,
      "eval_steps_per_second": 4.471,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 160.92691040039062,
      "learning_rate": 1.3180194846605365e-05,
      "loss": 0.3205,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.864,
      "eval_loss": 0.4371786415576935,
      "eval_runtime": 13.0617,
      "eval_samples_per_second": 287.099,
      "eval_steps_per_second": 4.517,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 193.30828857421875,
      "learning_rate": 3.425421036992097e-06,
      "loss": 0.2985,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8722666666666666,
      "eval_loss": 0.41767585277557373,
      "eval_runtime": 13.9694,
      "eval_samples_per_second": 268.443,
      "eval_steps_per_second": 4.224,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 97.36902618408203,
      "learning_rate": 0.0,
      "loss": 0.2852,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8728,
      "eval_loss": 0.42242714762687683,
      "eval_runtime": 13.4989,
      "eval_samples_per_second": 277.8,
      "eval_steps_per_second": 4.371,
      "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
}