trainer_state.json

{
  "best_metric": 0.7490636704119851,
  "best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-mobile-eye-tracking-dataset-v2/checkpoint-117",
  "epoch": 2.9433962264150946,
  "eval_steps": 500,
  "global_step": 117,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.25157232704402516,
      "grad_norm": 24.5244197845459,
      "learning_rate": 4.166666666666667e-05,
      "loss": 2.0156,
      "step": 10
    },
    {
      "epoch": 0.5031446540880503,
      "grad_norm": 23.752273559570312,
      "learning_rate": 4.6190476190476194e-05,
      "loss": 1.6205,
      "step": 20
    },
    {
      "epoch": 0.7547169811320755,
      "grad_norm": 27.802560806274414,
      "learning_rate": 4.1428571428571437e-05,
      "loss": 1.2929,
      "step": 30
    },
    {
      "epoch": 0.9811320754716981,
      "eval_accuracy": 0.6404494382022472,
      "eval_loss": 1.0328161716461182,
      "eval_runtime": 11.3947,
      "eval_samples_per_second": 23.432,
      "eval_steps_per_second": 0.79,
      "step": 39
    },
    {
      "epoch": 1.0062893081761006,
      "grad_norm": 40.85542297363281,
      "learning_rate": 3.6666666666666666e-05,
      "loss": 1.2308,
      "step": 40
    },
    {
      "epoch": 1.2578616352201257,
      "grad_norm": 36.25816345214844,
      "learning_rate": 3.19047619047619e-05,
      "loss": 1.1175,
      "step": 50
    },
    {
      "epoch": 1.509433962264151,
      "grad_norm": 27.19416046142578,
      "learning_rate": 2.714285714285714e-05,
      "loss": 1.0161,
      "step": 60
    },
    {
      "epoch": 1.7610062893081762,
      "grad_norm": 31.43399429321289,
      "learning_rate": 2.2380952380952384e-05,
      "loss": 0.9974,
      "step": 70
    },
    {
      "epoch": 1.9874213836477987,
      "eval_accuracy": 0.7415730337078652,
      "eval_loss": 0.7794644236564636,
      "eval_runtime": 10.2174,
      "eval_samples_per_second": 26.132,
      "eval_steps_per_second": 0.881,
      "step": 79
    },
    {
      "epoch": 2.0125786163522013,
      "grad_norm": 43.6927490234375,
      "learning_rate": 1.761904761904762e-05,
      "loss": 0.9507,
      "step": 80
    },
    {
      "epoch": 2.2641509433962264,
      "grad_norm": 43.01408386230469,
      "learning_rate": 1.2857142857142857e-05,
      "loss": 0.9205,
      "step": 90
    },
    {
      "epoch": 2.5157232704402515,
      "grad_norm": 33.98514175415039,
      "learning_rate": 8.095238095238097e-06,
      "loss": 0.9188,
      "step": 100
    },
    {
      "epoch": 2.767295597484277,
      "grad_norm": 33.26156997680664,
      "learning_rate": 3.3333333333333333e-06,
      "loss": 0.9114,
      "step": 110
    },
    {
      "epoch": 2.9433962264150946,
      "eval_accuracy": 0.7490636704119851,
      "eval_loss": 0.7275992035865784,
      "eval_runtime": 21.3079,
      "eval_samples_per_second": 12.531,
      "eval_steps_per_second": 0.422,
      "step": 117
    },
    {
      "epoch": 2.9433962264150946,
      "step": 117,
      "total_flos": 3.713320402332549e+17,
      "train_loss": 1.1595602361564963,
      "train_runtime": 718.6995,
      "train_samples_per_second": 21.155,
      "train_steps_per_second": 0.163
    }
  ],
  "logging_steps": 10,
  "max_steps": 117,
  "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": true
      },
      "attributes": {}
    }
  },
  "total_flos": 3.713320402332549e+17,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}