trainer_state.json

{
  "best_metric": 0.8938666666666667,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0109/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": 43.900325775146484,
      "learning_rate": 0.000288581929876693,
      "loss": 1.5081,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8325333333333333,
      "eval_loss": 0.5385944843292236,
      "eval_runtime": 14.3351,
      "eval_samples_per_second": 261.595,
      "eval_steps_per_second": 4.116,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 30.334033966064453,
      "learning_rate": 0.00025606601717798207,
      "loss": 0.4146,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8616,
      "eval_loss": 0.44018468260765076,
      "eval_runtime": 14.47,
      "eval_samples_per_second": 259.156,
      "eval_steps_per_second": 4.077,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 39.28211212158203,
      "learning_rate": 0.00020740251485476345,
      "loss": 0.2544,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8744,
      "eval_loss": 0.39958643913269043,
      "eval_runtime": 14.471,
      "eval_samples_per_second": 259.139,
      "eval_steps_per_second": 4.077,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 39.97348403930664,
      "learning_rate": 0.00015,
      "loss": 0.1568,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8850666666666667,
      "eval_loss": 0.39562779664993286,
      "eval_runtime": 14.3664,
      "eval_samples_per_second": 261.026,
      "eval_steps_per_second": 4.107,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 50.05442428588867,
      "learning_rate": 9.259748514523653e-05,
      "loss": 0.0889,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8888,
      "eval_loss": 0.4099200963973999,
      "eval_runtime": 14.6061,
      "eval_samples_per_second": 256.742,
      "eval_steps_per_second": 4.039,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 27.23273468017578,
      "learning_rate": 4.3933982822017876e-05,
      "loss": 0.0595,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.892,
      "eval_loss": 0.3928779065608978,
      "eval_runtime": 14.4376,
      "eval_samples_per_second": 259.738,
      "eval_steps_per_second": 4.087,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 26.5945987701416,
      "learning_rate": 1.1418070123306989e-05,
      "loss": 0.0366,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8938666666666667,
      "eval_loss": 0.38743671774864197,
      "eval_runtime": 14.4293,
      "eval_samples_per_second": 259.887,
      "eval_steps_per_second": 4.089,
      "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
}