trainer_state.json

{
  "best_metric": 0.9029333333333334,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0515/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": 151.1251220703125,
      "learning_rate": 7e-05,
      "loss": 3.2715,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6776,
      "eval_loss": 1.5875544548034668,
      "eval_runtime": 7.708,
      "eval_samples_per_second": 486.507,
      "eval_steps_per_second": 7.654,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 81.64579772949219,
      "learning_rate": 7e-05,
      "loss": 0.8587,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8437333333333333,
      "eval_loss": 0.542361319065094,
      "eval_runtime": 7.8773,
      "eval_samples_per_second": 476.052,
      "eval_steps_per_second": 7.49,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 78.3969955444336,
      "learning_rate": 7e-05,
      "loss": 0.4428,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8669333333333333,
      "eval_loss": 0.4561280906200409,
      "eval_runtime": 7.7675,
      "eval_samples_per_second": 482.783,
      "eval_steps_per_second": 7.596,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 158.80564880371094,
      "learning_rate": 7e-05,
      "loss": 0.2989,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.892,
      "eval_loss": 0.38159096240997314,
      "eval_runtime": 7.6906,
      "eval_samples_per_second": 487.607,
      "eval_steps_per_second": 7.672,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 135.48756408691406,
      "learning_rate": 7e-05,
      "loss": 0.2016,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8893333333333333,
      "eval_loss": 0.38033944368362427,
      "eval_runtime": 7.6691,
      "eval_samples_per_second": 488.973,
      "eval_steps_per_second": 7.693,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 144.16793823242188,
      "learning_rate": 7e-05,
      "loss": 0.1503,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8997333333333334,
      "eval_loss": 0.35698026418685913,
      "eval_runtime": 7.6862,
      "eval_samples_per_second": 487.884,
      "eval_steps_per_second": 7.676,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 171.06234741210938,
      "learning_rate": 7e-05,
      "loss": 0.1027,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9029333333333334,
      "eval_loss": 0.35705485939979553,
      "eval_runtime": 7.6156,
      "eval_samples_per_second": 492.413,
      "eval_steps_per_second": 7.747,
      "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
}