trainer_state.json

{
  "best_metric": 0.8682666666666666,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0471/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": 140.11019897460938,
      "learning_rate": 0.0001,
      "loss": 2.9193,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6858666666666666,
      "eval_loss": 1.1649166345596313,
      "eval_runtime": 15.0511,
      "eval_samples_per_second": 249.152,
      "eval_steps_per_second": 3.92,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 83.1082763671875,
      "learning_rate": 0.0001,
      "loss": 0.8324,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8141333333333334,
      "eval_loss": 0.6051420569419861,
      "eval_runtime": 15.1315,
      "eval_samples_per_second": 247.827,
      "eval_steps_per_second": 3.899,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 107.32686614990234,
      "learning_rate": 0.0001,
      "loss": 0.5147,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8384,
      "eval_loss": 0.5195069909095764,
      "eval_runtime": 15.029,
      "eval_samples_per_second": 249.518,
      "eval_steps_per_second": 3.926,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 52.05335998535156,
      "learning_rate": 0.0001,
      "loss": 0.3751,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8562666666666666,
      "eval_loss": 0.4621216058731079,
      "eval_runtime": 15.0123,
      "eval_samples_per_second": 249.795,
      "eval_steps_per_second": 3.93,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 91.27769470214844,
      "learning_rate": 0.0001,
      "loss": 0.2929,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8621333333333333,
      "eval_loss": 0.45018431544303894,
      "eval_runtime": 15.0532,
      "eval_samples_per_second": 249.116,
      "eval_steps_per_second": 3.919,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 86.88874053955078,
      "learning_rate": 0.0001,
      "loss": 0.2277,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8677333333333334,
      "eval_loss": 0.4584139585494995,
      "eval_runtime": 14.851,
      "eval_samples_per_second": 252.508,
      "eval_steps_per_second": 3.973,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 49.56699752807617,
      "learning_rate": 0.0001,
      "loss": 0.1797,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8682666666666666,
      "eval_loss": 0.4332180917263031,
      "eval_runtime": 15.0536,
      "eval_samples_per_second": 249.109,
      "eval_steps_per_second": 3.919,
      "step": 2331
    }
  ],
  "logging_steps": 500,
  "max_steps": 2331,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "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": 5.7235004676864e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}