checkpoint-3750/trainer_state.json

{
  "best_metric": 0.18133555352687836,
  "best_model_checkpoint": "fairface_gender_image_detection/checkpoint-3750",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 3750,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.26666666666666666,
      "grad_norm": 3.194249391555786,
      "learning_rate": 8.273542600896861e-06,
      "loss": 0.3452,
      "step": 500
    },
    {
      "epoch": 0.5333333333333333,
      "grad_norm": 2.3906941413879395,
      "learning_rate": 7.4663677130044845e-06,
      "loss": 0.1955,
      "step": 1000
    },
    {
      "epoch": 0.8,
      "grad_norm": 3.958942174911499,
      "learning_rate": 6.6591928251121084e-06,
      "loss": 0.1709,
      "step": 1500
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.915,
      "eval_loss": 0.21168974041938782,
      "eval_model_preparation_time": 0.0041,
      "eval_runtime": 890.2454,
      "eval_samples_per_second": 11.233,
      "eval_steps_per_second": 1.404,
      "step": 1875
    },
    {
      "epoch": 1.0666666666666667,
      "grad_norm": 6.014321804046631,
      "learning_rate": 5.852017937219731e-06,
      "loss": 0.154,
      "step": 2000
    },
    {
      "epoch": 1.3333333333333333,
      "grad_norm": 10.029616355895996,
      "learning_rate": 5.044843049327354e-06,
      "loss": 0.1161,
      "step": 2500
    },
    {
      "epoch": 1.6,
      "grad_norm": 2.819401979446411,
      "learning_rate": 4.237668161434978e-06,
      "loss": 0.1218,
      "step": 3000
    },
    {
      "epoch": 1.8666666666666667,
      "grad_norm": 1.9613158702850342,
      "learning_rate": 3.430493273542601e-06,
      "loss": 0.1212,
      "step": 3500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9342,
      "eval_loss": 0.18133555352687836,
      "eval_model_preparation_time": 0.0041,
      "eval_runtime": 894.9262,
      "eval_samples_per_second": 11.174,
      "eval_steps_per_second": 1.397,
      "step": 3750
    }
  ],
  "logging_steps": 500,
  "max_steps": 5625,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "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": 9.29903875375104e+18,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}