trainer_state.json

{
  "best_metric": 0.888,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0589/checkpoints/checkpoint-1998",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 1998,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 153.33596801757812,
      "learning_rate": 9.504844339512095e-05,
      "loss": 2.7022,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7738666666666667,
      "eval_loss": 0.8809183835983276,
      "eval_runtime": 7.9124,
      "eval_samples_per_second": 473.941,
      "eval_steps_per_second": 7.457,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 69.81400299072266,
      "learning_rate": 8.117449009293668e-05,
      "loss": 0.6955,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8376,
      "eval_loss": 0.547929048538208,
      "eval_runtime": 7.9291,
      "eval_samples_per_second": 472.942,
      "eval_steps_per_second": 7.441,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 178.31866455078125,
      "learning_rate": 6.112604669781572e-05,
      "loss": 0.4424,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8653333333333333,
      "eval_loss": 0.44758105278015137,
      "eval_runtime": 7.8857,
      "eval_samples_per_second": 475.546,
      "eval_steps_per_second": 7.482,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 194.490234375,
      "learning_rate": 3.887395330218429e-05,
      "loss": 0.3187,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8778666666666667,
      "eval_loss": 0.4064708352088928,
      "eval_runtime": 7.8665,
      "eval_samples_per_second": 476.706,
      "eval_steps_per_second": 7.5,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 169.18296813964844,
      "learning_rate": 1.8825509907063327e-05,
      "loss": 0.2507,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8837333333333334,
      "eval_loss": 0.38955172896385193,
      "eval_runtime": 7.8373,
      "eval_samples_per_second": 478.482,
      "eval_steps_per_second": 7.528,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 114.50824737548828,
      "learning_rate": 4.951556604879048e-06,
      "loss": 0.2211,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.888,
      "eval_loss": 0.3814295530319214,
      "eval_runtime": 7.8669,
      "eval_samples_per_second": 476.678,
      "eval_steps_per_second": 7.5,
      "step": 1998
    }
  ],
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 4.9058575437312e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}