{
  "best_metric": 0.6739130434782609,
  "best_model_checkpoint": "swinv2-tiny-patch4-window8-256-DMAE-3e-5\\checkpoint-94",
  "epoch": 34.285714285714285,
  "eval_steps": 500,
  "global_step": 120,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.86,
      "eval_accuracy": 0.10869565217391304,
      "eval_loss": 1.6337018013000488,
      "eval_runtime": 0.8832,
      "eval_samples_per_second": 52.083,
      "eval_steps_per_second": 3.397,
      "step": 3
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.10869565217391304,
      "eval_loss": 1.622289776802063,
      "eval_runtime": 0.9452,
      "eval_samples_per_second": 48.666,
      "eval_steps_per_second": 3.174,
      "step": 7
    },
    {
      "epoch": 2.86,
      "learning_rate": 2.5e-05,
      "loss": 1.6298,
      "step": 10
    },
    {
      "epoch": 2.86,
      "eval_accuracy": 0.10869565217391304,
      "eval_loss": 1.5971730947494507,
      "eval_runtime": 0.6641,
      "eval_samples_per_second": 69.268,
      "eval_steps_per_second": 4.517,
      "step": 10
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.10869565217391304,
      "eval_loss": 1.5040873289108276,
      "eval_runtime": 0.7672,
      "eval_samples_per_second": 59.96,
      "eval_steps_per_second": 3.91,
      "step": 14
    },
    {
      "epoch": 4.86,
      "eval_accuracy": 0.10869565217391304,
      "eval_loss": 1.4149224758148193,
      "eval_runtime": 0.8815,
      "eval_samples_per_second": 52.182,
      "eval_steps_per_second": 3.403,
      "step": 17
    },
    {
      "epoch": 5.71,
      "learning_rate": 2.777777777777778e-05,
      "loss": 1.4858,
      "step": 20
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.45652173913043476,
      "eval_loss": 1.3036504983901978,
      "eval_runtime": 0.6381,
      "eval_samples_per_second": 72.084,
      "eval_steps_per_second": 4.701,
      "step": 21
    },
    {
      "epoch": 6.86,
      "eval_accuracy": 0.45652173913043476,
      "eval_loss": 1.2491180896759033,
      "eval_runtime": 0.6833,
      "eval_samples_per_second": 67.325,
      "eval_steps_per_second": 4.391,
      "step": 24
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.45652173913043476,
      "eval_loss": 1.2219406366348267,
      "eval_runtime": 0.6631,
      "eval_samples_per_second": 69.366,
      "eval_steps_per_second": 4.524,
      "step": 28
    },
    {
      "epoch": 8.57,
      "learning_rate": 2.5e-05,
      "loss": 1.2337,
      "step": 30
    },
    {
      "epoch": 8.86,
      "eval_accuracy": 0.45652173913043476,
      "eval_loss": 1.2146371603012085,
      "eval_runtime": 0.6597,
      "eval_samples_per_second": 69.732,
      "eval_steps_per_second": 4.548,
      "step": 31
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.45652173913043476,
      "eval_loss": 1.214502215385437,
      "eval_runtime": 0.6581,
      "eval_samples_per_second": 69.893,
      "eval_steps_per_second": 4.558,
      "step": 35
    },
    {
      "epoch": 10.86,
      "eval_accuracy": 0.45652173913043476,
      "eval_loss": 1.211630940437317,
      "eval_runtime": 0.6381,
      "eval_samples_per_second": 72.083,
      "eval_steps_per_second": 4.701,
      "step": 38
    },
    {
      "epoch": 11.43,
      "learning_rate": 2.222222222222222e-05,
      "loss": 1.1726,
      "step": 40
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.45652173913043476,
      "eval_loss": 1.2064777612686157,
      "eval_runtime": 0.6501,
      "eval_samples_per_second": 70.753,
      "eval_steps_per_second": 4.614,
      "step": 42
    },
    {
      "epoch": 12.86,
      "eval_accuracy": 0.45652173913043476,
      "eval_loss": 1.201141357421875,
      "eval_runtime": 0.6601,
      "eval_samples_per_second": 69.681,
      "eval_steps_per_second": 4.544,
      "step": 45
    },
    {
      "epoch": 14.0,
      "eval_accuracy": 0.45652173913043476,
      "eval_loss": 1.1940866708755493,
      "eval_runtime": 0.6627,
      "eval_samples_per_second": 69.413,
      "eval_steps_per_second": 4.527,
      "step": 49
    },
    {
      "epoch": 14.29,
      "learning_rate": 1.9444444444444445e-05,
      "loss": 1.1953,
      "step": 50
    },
    {
      "epoch": 14.86,
      "eval_accuracy": 0.45652173913043476,
      "eval_loss": 1.18779456615448,
      "eval_runtime": 0.695,
      "eval_samples_per_second": 66.19,
      "eval_steps_per_second": 4.317,
      "step": 52
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.45652173913043476,
      "eval_loss": 1.1770756244659424,
      "eval_runtime": 0.6902,
      "eval_samples_per_second": 66.652,
      "eval_steps_per_second": 4.347,
      "step": 56
    },
    {
      "epoch": 16.86,
      "eval_accuracy": 0.45652173913043476,
      "eval_loss": 1.1677342653274536,
      "eval_runtime": 0.6555,
      "eval_samples_per_second": 70.174,
      "eval_steps_per_second": 4.577,
      "step": 59
    },
    {
      "epoch": 17.14,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 1.1602,
      "step": 60
    },
    {
      "epoch": 18.0,
      "eval_accuracy": 0.45652173913043476,
      "eval_loss": 1.1564418077468872,
      "eval_runtime": 0.6792,
      "eval_samples_per_second": 67.731,
      "eval_steps_per_second": 4.417,
      "step": 63
    },
    {
      "epoch": 18.86,
      "eval_accuracy": 0.45652173913043476,
      "eval_loss": 1.1501902341842651,
      "eval_runtime": 0.6591,
      "eval_samples_per_second": 69.795,
      "eval_steps_per_second": 4.552,
      "step": 66
    },
    {
      "epoch": 20.0,
      "learning_rate": 1.388888888888889e-05,
      "loss": 1.1415,
      "step": 70
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 0.45652173913043476,
      "eval_loss": 1.1226557493209839,
      "eval_runtime": 0.6682,
      "eval_samples_per_second": 68.846,
      "eval_steps_per_second": 4.49,
      "step": 70
    },
    {
      "epoch": 20.86,
      "eval_accuracy": 0.4782608695652174,
      "eval_loss": 1.101697564125061,
      "eval_runtime": 0.6411,
      "eval_samples_per_second": 71.747,
      "eval_steps_per_second": 4.679,
      "step": 73
    },
    {
      "epoch": 22.0,
      "eval_accuracy": 0.5869565217391305,
      "eval_loss": 1.0871663093566895,
      "eval_runtime": 0.6591,
      "eval_samples_per_second": 69.787,
      "eval_steps_per_second": 4.551,
      "step": 77
    },
    {
      "epoch": 22.86,
      "learning_rate": 1.111111111111111e-05,
      "loss": 1.0949,
      "step": 80
    },
    {
      "epoch": 22.86,
      "eval_accuracy": 0.5869565217391305,
      "eval_loss": 1.0795304775238037,
      "eval_runtime": 0.6621,
      "eval_samples_per_second": 69.471,
      "eval_steps_per_second": 4.531,
      "step": 80
    },
    {
      "epoch": 24.0,
      "eval_accuracy": 0.6086956521739131,
      "eval_loss": 1.0581977367401123,
      "eval_runtime": 0.6631,
      "eval_samples_per_second": 69.366,
      "eval_steps_per_second": 4.524,
      "step": 84
    },
    {
      "epoch": 24.86,
      "eval_accuracy": 0.6086956521739131,
      "eval_loss": 1.0401618480682373,
      "eval_runtime": 0.7072,
      "eval_samples_per_second": 65.047,
      "eval_steps_per_second": 4.242,
      "step": 87
    },
    {
      "epoch": 25.71,
      "learning_rate": 8.333333333333334e-06,
      "loss": 1.0559,
      "step": 90
    },
    {
      "epoch": 26.0,
      "eval_accuracy": 0.6521739130434783,
      "eval_loss": 1.0186063051223755,
      "eval_runtime": 0.6467,
      "eval_samples_per_second": 71.135,
      "eval_steps_per_second": 4.639,
      "step": 91
    },
    {
      "epoch": 26.86,
      "eval_accuracy": 0.6739130434782609,
      "eval_loss": 1.01261305809021,
      "eval_runtime": 0.6762,
      "eval_samples_per_second": 68.031,
      "eval_steps_per_second": 4.437,
      "step": 94
    },
    {
      "epoch": 28.0,
      "eval_accuracy": 0.6739130434782609,
      "eval_loss": 1.0032662153244019,
      "eval_runtime": 0.6702,
      "eval_samples_per_second": 68.639,
      "eval_steps_per_second": 4.476,
      "step": 98
    },
    {
      "epoch": 28.57,
      "learning_rate": 5.555555555555555e-06,
      "loss": 1.0423,
      "step": 100
    },
    {
      "epoch": 28.86,
      "eval_accuracy": 0.6739130434782609,
      "eval_loss": 0.9848656058311462,
      "eval_runtime": 0.6762,
      "eval_samples_per_second": 68.029,
      "eval_steps_per_second": 4.437,
      "step": 101
    },
    {
      "epoch": 30.0,
      "eval_accuracy": 0.6739130434782609,
      "eval_loss": 0.9669570922851562,
      "eval_runtime": 0.6637,
      "eval_samples_per_second": 69.314,
      "eval_steps_per_second": 4.52,
      "step": 105
    },
    {
      "epoch": 30.86,
      "eval_accuracy": 0.6739130434782609,
      "eval_loss": 0.9630699753761292,
      "eval_runtime": 0.8147,
      "eval_samples_per_second": 56.462,
      "eval_steps_per_second": 3.682,
      "step": 108
    },
    {
      "epoch": 31.43,
      "learning_rate": 2.7777777777777775e-06,
      "loss": 1.0002,
      "step": 110
    },
    {
      "epoch": 32.0,
      "eval_accuracy": 0.6521739130434783,
      "eval_loss": 0.9599204659461975,
      "eval_runtime": 0.6631,
      "eval_samples_per_second": 69.366,
      "eval_steps_per_second": 4.524,
      "step": 112
    },
    {
      "epoch": 32.86,
      "eval_accuracy": 0.6521739130434783,
      "eval_loss": 0.9586823582649231,
      "eval_runtime": 0.6661,
      "eval_samples_per_second": 69.054,
      "eval_steps_per_second": 4.503,
      "step": 115
    },
    {
      "epoch": 34.0,
      "eval_accuracy": 0.6521739130434783,
      "eval_loss": 0.9573362469673157,
      "eval_runtime": 0.6584,
      "eval_samples_per_second": 69.871,
      "eval_steps_per_second": 4.557,
      "step": 119
    },
    {
      "epoch": 34.29,
      "learning_rate": 0.0,
      "loss": 1.0077,
      "step": 120
    },
    {
      "epoch": 34.29,
      "eval_accuracy": 0.6521739130434783,
      "eval_loss": 0.9570884704589844,
      "eval_runtime": 0.6737,
      "eval_samples_per_second": 68.281,
      "eval_steps_per_second": 4.453,
      "step": 120
    },
    {
      "epoch": 34.29,
      "step": 120,
      "total_flos": 2.3770905934823424e+17,
      "train_loss": 1.18499755859375,
      "train_runtime": 199.3828,
      "train_samples_per_second": 42.732,
      "train_steps_per_second": 0.602
    }
  ],
  "logging_steps": 10,
  "max_steps": 120,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 40,
  "save_steps": 500,
  "total_flos": 2.3770905934823424e+17,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}