trainer_state.json

{
  "best_metric": 0.9075907590759076,
  "best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-lungs-disease/checkpoint-105",
  "epoch": 4.883720930232558,
  "eval_steps": 500,
  "global_step": 105,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.47,
      "grad_norm": 6.489995002746582,
      "learning_rate": 4.545454545454546e-05,
      "loss": 1.4339,
      "step": 10
    },
    {
      "epoch": 0.93,
      "grad_norm": 6.888110637664795,
      "learning_rate": 4.5212765957446815e-05,
      "loss": 0.814,
      "step": 20
    },
    {
      "epoch": 0.98,
      "eval_accuracy": 0.7491749174917491,
      "eval_loss": 0.5313370227813721,
      "eval_runtime": 66.4067,
      "eval_samples_per_second": 4.563,
      "eval_steps_per_second": 0.151,
      "step": 21
    },
    {
      "epoch": 1.4,
      "grad_norm": 15.805723190307617,
      "learning_rate": 3.9893617021276594e-05,
      "loss": 0.5569,
      "step": 30
    },
    {
      "epoch": 1.86,
      "grad_norm": 13.49924087524414,
      "learning_rate": 3.4574468085106386e-05,
      "loss": 0.4444,
      "step": 40
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8910891089108911,
      "eval_loss": 0.3199963867664337,
      "eval_runtime": 66.8668,
      "eval_samples_per_second": 4.531,
      "eval_steps_per_second": 0.15,
      "step": 43
    },
    {
      "epoch": 2.33,
      "grad_norm": 8.2146577835083,
      "learning_rate": 2.925531914893617e-05,
      "loss": 0.3767,
      "step": 50
    },
    {
      "epoch": 2.79,
      "grad_norm": 12.676996231079102,
      "learning_rate": 2.393617021276596e-05,
      "loss": 0.3322,
      "step": 60
    },
    {
      "epoch": 2.98,
      "eval_accuracy": 0.8910891089108911,
      "eval_loss": 0.31477591395378113,
      "eval_runtime": 66.222,
      "eval_samples_per_second": 4.576,
      "eval_steps_per_second": 0.151,
      "step": 64
    },
    {
      "epoch": 3.26,
      "grad_norm": 9.924786567687988,
      "learning_rate": 1.8617021276595745e-05,
      "loss": 0.2842,
      "step": 70
    },
    {
      "epoch": 3.72,
      "grad_norm": 8.528675079345703,
      "learning_rate": 1.3297872340425532e-05,
      "loss": 0.2975,
      "step": 80
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8976897689768977,
      "eval_loss": 0.28364694118499756,
      "eval_runtime": 66.122,
      "eval_samples_per_second": 4.582,
      "eval_steps_per_second": 0.151,
      "step": 86
    },
    {
      "epoch": 4.19,
      "grad_norm": 16.571998596191406,
      "learning_rate": 7.97872340425532e-06,
      "loss": 0.2764,
      "step": 90
    },
    {
      "epoch": 4.65,
      "grad_norm": 6.849820613861084,
      "learning_rate": 2.6595744680851065e-06,
      "loss": 0.254,
      "step": 100
    },
    {
      "epoch": 4.88,
      "eval_accuracy": 0.9075907590759076,
      "eval_loss": 0.24336053431034088,
      "eval_runtime": 66.2106,
      "eval_samples_per_second": 4.576,
      "eval_steps_per_second": 0.151,
      "step": 105
    },
    {
      "epoch": 4.88,
      "step": 105,
      "total_flos": 3.313083915385897e+17,
      "train_loss": 0.4951313120978219,
      "train_runtime": 9028.2693,
      "train_samples_per_second": 1.509,
      "train_steps_per_second": 0.012
    }
  ],
  "logging_steps": 10,
  "max_steps": 105,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 3.313083915385897e+17,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}