trainer_state.json

{
  "best_metric": 0.8872,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0064/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": 121.43282318115234,
      "learning_rate": 7e-05,
      "loss": 3.3039,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6168,
      "eval_loss": 1.715225338935852,
      "eval_runtime": 13.4605,
      "eval_samples_per_second": 278.593,
      "eval_steps_per_second": 4.383,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 113.20372009277344,
      "learning_rate": 7e-05,
      "loss": 1.0426,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8066666666666666,
      "eval_loss": 0.70968097448349,
      "eval_runtime": 13.2024,
      "eval_samples_per_second": 284.04,
      "eval_steps_per_second": 4.469,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 82.30632781982422,
      "learning_rate": 7e-05,
      "loss": 0.5956,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8392,
      "eval_loss": 0.5469499826431274,
      "eval_runtime": 13.5271,
      "eval_samples_per_second": 277.222,
      "eval_steps_per_second": 4.362,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 176.51918029785156,
      "learning_rate": 7e-05,
      "loss": 0.4355,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8610666666666666,
      "eval_loss": 0.4657478630542755,
      "eval_runtime": 13.3887,
      "eval_samples_per_second": 280.088,
      "eval_steps_per_second": 4.407,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 145.4884490966797,
      "learning_rate": 7e-05,
      "loss": 0.3414,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8696,
      "eval_loss": 0.42032772302627563,
      "eval_runtime": 13.3861,
      "eval_samples_per_second": 280.141,
      "eval_steps_per_second": 4.408,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 60.22314453125,
      "learning_rate": 7e-05,
      "loss": 0.2653,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8789333333333333,
      "eval_loss": 0.40077316761016846,
      "eval_runtime": 14.0178,
      "eval_samples_per_second": 267.517,
      "eval_steps_per_second": 4.209,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 95.4765625,
      "learning_rate": 7e-05,
      "loss": 0.2055,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8805333333333333,
      "eval_loss": 0.4131264388561249,
      "eval_runtime": 13.8922,
      "eval_samples_per_second": 269.936,
      "eval_steps_per_second": 4.247,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 96.52843475341797,
      "learning_rate": 7e-05,
      "loss": 0.1788,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8872,
      "eval_loss": 0.3821322023868561,
      "eval_runtime": 13.962,
      "eval_samples_per_second": 268.587,
      "eval_steps_per_second": 4.226,
      "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
}