trainer_state.json

{
  "best_metric": 0.8997333333333334,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0742/checkpoints/checkpoint-2664",
  "epoch": 8.0,
  "eval_steps": 500,
  "global_step": 2664,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 70.2479019165039,
      "learning_rate": 9e-05,
      "loss": 2.6956,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.788,
      "eval_loss": 0.8131048083305359,
      "eval_runtime": 7.3188,
      "eval_samples_per_second": 512.379,
      "eval_steps_per_second": 8.061,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 139.388916015625,
      "learning_rate": 9e-05,
      "loss": 0.5933,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8565333333333334,
      "eval_loss": 0.46711465716362,
      "eval_runtime": 7.3261,
      "eval_samples_per_second": 511.871,
      "eval_steps_per_second": 8.053,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 84.93392944335938,
      "learning_rate": 9e-05,
      "loss": 0.3547,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8770666666666667,
      "eval_loss": 0.3996787369251251,
      "eval_runtime": 7.2063,
      "eval_samples_per_second": 520.379,
      "eval_steps_per_second": 8.187,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 113.6795654296875,
      "learning_rate": 9e-05,
      "loss": 0.2501,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8893333333333333,
      "eval_loss": 0.35488182306289673,
      "eval_runtime": 7.6304,
      "eval_samples_per_second": 491.453,
      "eval_steps_per_second": 7.732,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 83.28839111328125,
      "learning_rate": 9e-05,
      "loss": 0.183,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8970666666666667,
      "eval_loss": 0.33955541253089905,
      "eval_runtime": 7.3433,
      "eval_samples_per_second": 510.667,
      "eval_steps_per_second": 8.035,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 119.84217834472656,
      "learning_rate": 9e-05,
      "loss": 0.1392,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8973333333333333,
      "eval_loss": 0.35965728759765625,
      "eval_runtime": 7.1857,
      "eval_samples_per_second": 521.868,
      "eval_steps_per_second": 8.211,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 43.34016036987305,
      "learning_rate": 9e-05,
      "loss": 0.1028,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8968,
      "eval_loss": 0.3685609996318817,
      "eval_runtime": 7.1392,
      "eval_samples_per_second": 525.271,
      "eval_steps_per_second": 8.264,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 117.14539337158203,
      "learning_rate": 9e-05,
      "loss": 0.0757,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8997333333333334,
      "eval_loss": 0.3570169508457184,
      "eval_runtime": 7.1838,
      "eval_samples_per_second": 522.011,
      "eval_steps_per_second": 8.213,
      "step": 2664
    }
  ],
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 6.5411433916416e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}