trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 200,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 618.759521484375,
      "learning_rate": 1e-05,
      "loss": 37.3934,
      "step": 100
    },
    {
      "epoch": 1.0,
      "eval_loss": 33.63319778442383,
      "eval_map": 0.1981,
      "eval_map_50": 0.3956,
      "eval_map_75": 0.1799,
      "eval_map_car": 0.2613,
      "eval_map_large": 0.7992,
      "eval_map_medium": 0.6173,
      "eval_map_person": 0.1349,
      "eval_map_small": 0.1285,
      "eval_mar_1": 0.0896,
      "eval_mar_10": 0.2453,
      "eval_mar_100": 0.2821,
      "eval_mar_100_car": 0.3641,
      "eval_mar_100_person": 0.2001,
      "eval_mar_large": 0.8354,
      "eval_mar_medium": 0.7135,
      "eval_mar_small": 0.2074,
      "eval_runtime": 54.9214,
      "eval_samples_per_second": 6.009,
      "eval_steps_per_second": 0.765,
      "step": 100
    },
    {
      "epoch": 2.0,
      "grad_norm": 113.26927947998047,
      "learning_rate": 1e-05,
      "loss": 29.0441,
      "step": 200
    },
    {
      "epoch": 2.0,
      "eval_loss": 31.006668090820312,
      "eval_map": 0.2061,
      "eval_map_50": 0.4076,
      "eval_map_75": 0.1946,
      "eval_map_car": 0.2709,
      "eval_map_large": 0.82,
      "eval_map_medium": 0.623,
      "eval_map_person": 0.1413,
      "eval_map_small": 0.1368,
      "eval_mar_1": 0.0921,
      "eval_mar_10": 0.2488,
      "eval_mar_100": 0.2856,
      "eval_mar_100_car": 0.3678,
      "eval_mar_100_person": 0.2035,
      "eval_mar_large": 0.8705,
      "eval_mar_medium": 0.7164,
      "eval_mar_small": 0.2105,
      "eval_runtime": 58.669,
      "eval_samples_per_second": 5.625,
      "eval_steps_per_second": 0.716,
      "step": 200
    },
    {
      "epoch": 2.0,
      "step": 200,
      "total_flos": 1.787388644818944e+17,
      "train_loss": 33.218726806640625,
      "train_runtime": 300.7183,
      "train_samples_per_second": 10.641,
      "train_steps_per_second": 0.665
    }
  ],
  "logging_steps": 500,
  "max_steps": 200,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "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": 1.787388644818944e+17,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}