trainer_state.json

{
  "best_metric": 0.8968,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0150/checkpoints/checkpoint-2997",
  "epoch": 9.0,
  "eval_steps": 500,
  "global_step": 2997,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 61.400169372558594,
      "learning_rate": 0.0003,
      "loss": 1.6039,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8261333333333334,
      "eval_loss": 0.5675727128982544,
      "eval_runtime": 9.209,
      "eval_samples_per_second": 407.21,
      "eval_steps_per_second": 6.407,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 48.256202697753906,
      "learning_rate": 0.0003,
      "loss": 0.447,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8642666666666666,
      "eval_loss": 0.4608747363090515,
      "eval_runtime": 9.2663,
      "eval_samples_per_second": 404.691,
      "eval_steps_per_second": 6.367,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 31.33218765258789,
      "learning_rate": 0.0003,
      "loss": 0.2782,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8813333333333333,
      "eval_loss": 0.41268372535705566,
      "eval_runtime": 9.0972,
      "eval_samples_per_second": 412.216,
      "eval_steps_per_second": 6.486,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 28.19144058227539,
      "learning_rate": 0.0003,
      "loss": 0.1885,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8826666666666667,
      "eval_loss": 0.40849393606185913,
      "eval_runtime": 9.0724,
      "eval_samples_per_second": 413.341,
      "eval_steps_per_second": 6.503,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 44.106693267822266,
      "learning_rate": 0.0003,
      "loss": 0.1406,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8824,
      "eval_loss": 0.4433552622795105,
      "eval_runtime": 9.0556,
      "eval_samples_per_second": 414.11,
      "eval_steps_per_second": 6.515,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 32.17176818847656,
      "learning_rate": 0.0003,
      "loss": 0.1094,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.892,
      "eval_loss": 0.41006290912628174,
      "eval_runtime": 9.0659,
      "eval_samples_per_second": 413.636,
      "eval_steps_per_second": 6.508,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 28.66417121887207,
      "learning_rate": 0.0003,
      "loss": 0.0885,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8928,
      "eval_loss": 0.4724870026111603,
      "eval_runtime": 9.0538,
      "eval_samples_per_second": 414.192,
      "eval_steps_per_second": 6.517,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 52.52780532836914,
      "learning_rate": 0.0003,
      "loss": 0.0761,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8882666666666666,
      "eval_loss": 0.4813437759876251,
      "eval_runtime": 9.1739,
      "eval_samples_per_second": 408.77,
      "eval_steps_per_second": 6.431,
      "step": 2664
    },
    {
      "epoch": 9.0,
      "grad_norm": 29.16042137145996,
      "learning_rate": 0.0003,
      "loss": 0.054,
      "step": 2997
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.8968,
      "eval_loss": 0.48102179169654846,
      "eval_runtime": 9.3446,
      "eval_samples_per_second": 401.302,
      "eval_steps_per_second": 6.314,
      "step": 2997
    }
  ],
  "logging_steps": 500,
  "max_steps": 2997,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9,
  "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": 7.3587863155968e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}