trainer_state.json

{
  "best_metric": 0.8688,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0746/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": 59.76859664916992,
      "learning_rate": 3e-05,
      "loss": 3.8357,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.43066666666666664,
      "eval_loss": 3.6950223445892334,
      "eval_runtime": 14.1482,
      "eval_samples_per_second": 265.052,
      "eval_steps_per_second": 4.17,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 74.01223754882812,
      "learning_rate": 3e-05,
      "loss": 3.015,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6069333333333333,
      "eval_loss": 2.2957568168640137,
      "eval_runtime": 14.0834,
      "eval_samples_per_second": 266.271,
      "eval_steps_per_second": 4.189,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 150.30279541015625,
      "learning_rate": 3e-05,
      "loss": 1.4154,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7653333333333333,
      "eval_loss": 0.9585574865341187,
      "eval_runtime": 14.0973,
      "eval_samples_per_second": 266.008,
      "eval_steps_per_second": 4.185,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 198.65127563476562,
      "learning_rate": 3e-05,
      "loss": 0.8002,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8042666666666667,
      "eval_loss": 0.6874225735664368,
      "eval_runtime": 14.0985,
      "eval_samples_per_second": 265.985,
      "eval_steps_per_second": 4.185,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 180.91651916503906,
      "learning_rate": 3e-05,
      "loss": 0.576,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8362666666666667,
      "eval_loss": 0.5552025437355042,
      "eval_runtime": 14.0897,
      "eval_samples_per_second": 266.152,
      "eval_steps_per_second": 4.187,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 164.390869140625,
      "learning_rate": 3e-05,
      "loss": 0.4529,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8466666666666667,
      "eval_loss": 0.4913294315338135,
      "eval_runtime": 14.1133,
      "eval_samples_per_second": 265.706,
      "eval_steps_per_second": 4.18,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 244.69390869140625,
      "learning_rate": 3e-05,
      "loss": 0.374,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8501333333333333,
      "eval_loss": 0.478765606880188,
      "eval_runtime": 14.1046,
      "eval_samples_per_second": 265.871,
      "eval_steps_per_second": 4.183,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 226.35818481445312,
      "learning_rate": 3e-05,
      "loss": 0.3001,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8688,
      "eval_loss": 0.4377187490463257,
      "eval_runtime": 14.0467,
      "eval_samples_per_second": 266.967,
      "eval_steps_per_second": 4.2,
      "step": 2664
    }
  ],
  "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": true
      },
      "attributes": {}
    }
  },
  "total_flos": 6.5411433916416e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}