{
  "best_metric": 0.6521739130434783,
  "best_model_checkpoint": "swinv2-tiny-patch4-window8-256-DMAE-U2\\checkpoint-63",
  "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": 0.4782608695652174,
      "eval_loss": 1.384318470954895,
      "eval_runtime": 0.7692,
      "eval_samples_per_second": 59.802,
      "eval_steps_per_second": 3.9,
      "step": 3
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.45652173913043476,
      "eval_loss": 1.3617327213287354,
      "eval_runtime": 0.7541,
      "eval_samples_per_second": 61.001,
      "eval_steps_per_second": 3.978,
      "step": 7
    },
    {
      "epoch": 2.86,
      "learning_rate": 3.37719298245614e-05,
      "loss": 1.3721,
      "step": 10
    },
    {
      "epoch": 2.86,
      "eval_accuracy": 0.45652173913043476,
      "eval_loss": 1.3210632801055908,
      "eval_runtime": 0.7782,
      "eval_samples_per_second": 59.112,
      "eval_steps_per_second": 3.855,
      "step": 10
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.45652173913043476,
      "eval_loss": 1.2598481178283691,
      "eval_runtime": 0.7401,
      "eval_samples_per_second": 62.152,
      "eval_steps_per_second": 4.053,
      "step": 14
    },
    {
      "epoch": 4.86,
      "eval_accuracy": 0.45652173913043476,
      "eval_loss": 1.2250741720199585,
      "eval_runtime": 0.8846,
      "eval_samples_per_second": 51.999,
      "eval_steps_per_second": 3.391,
      "step": 17
    },
    {
      "epoch": 5.71,
      "learning_rate": 3.070175438596491e-05,
      "loss": 1.25,
      "step": 20
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.45652173913043476,
      "eval_loss": 1.209437608718872,
      "eval_runtime": 0.7856,
      "eval_samples_per_second": 58.558,
      "eval_steps_per_second": 3.819,
      "step": 21
    },
    {
      "epoch": 6.86,
      "eval_accuracy": 0.45652173913043476,
      "eval_loss": 1.2126904726028442,
      "eval_runtime": 0.7803,
      "eval_samples_per_second": 58.952,
      "eval_steps_per_second": 3.845,
      "step": 24
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.45652173913043476,
      "eval_loss": 1.2118207216262817,
      "eval_runtime": 0.7538,
      "eval_samples_per_second": 61.026,
      "eval_steps_per_second": 3.98,
      "step": 28
    },
    {
      "epoch": 8.57,
      "learning_rate": 2.763157894736842e-05,
      "loss": 1.1902,
      "step": 30
    },
    {
      "epoch": 8.86,
      "eval_accuracy": 0.45652173913043476,
      "eval_loss": 1.203112006187439,
      "eval_runtime": 0.776,
      "eval_samples_per_second": 59.278,
      "eval_steps_per_second": 3.866,
      "step": 31
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.45652173913043476,
      "eval_loss": 1.1936233043670654,
      "eval_runtime": 0.746,
      "eval_samples_per_second": 61.659,
      "eval_steps_per_second": 4.021,
      "step": 35
    },
    {
      "epoch": 10.86,
      "eval_accuracy": 0.45652173913043476,
      "eval_loss": 1.1829485893249512,
      "eval_runtime": 0.794,
      "eval_samples_per_second": 57.938,
      "eval_steps_per_second": 3.779,
      "step": 38
    },
    {
      "epoch": 11.43,
      "learning_rate": 2.4561403508771925e-05,
      "loss": 1.1472,
      "step": 40
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.45652173913043476,
      "eval_loss": 1.156883955001831,
      "eval_runtime": 0.744,
      "eval_samples_per_second": 61.827,
      "eval_steps_per_second": 4.032,
      "step": 42
    },
    {
      "epoch": 12.86,
      "eval_accuracy": 0.45652173913043476,
      "eval_loss": 1.1431748867034912,
      "eval_runtime": 0.7287,
      "eval_samples_per_second": 63.129,
      "eval_steps_per_second": 4.117,
      "step": 45
    },
    {
      "epoch": 14.0,
      "eval_accuracy": 0.4782608695652174,
      "eval_loss": 1.1357367038726807,
      "eval_runtime": 0.7642,
      "eval_samples_per_second": 60.196,
      "eval_steps_per_second": 3.926,
      "step": 49
    },
    {
      "epoch": 14.29,
      "learning_rate": 2.1491228070175438e-05,
      "loss": 1.1495,
      "step": 50
    },
    {
      "epoch": 14.86,
      "eval_accuracy": 0.5,
      "eval_loss": 1.1177672147750854,
      "eval_runtime": 0.7578,
      "eval_samples_per_second": 60.705,
      "eval_steps_per_second": 3.959,
      "step": 52
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.5217391304347826,
      "eval_loss": 1.0903263092041016,
      "eval_runtime": 0.77,
      "eval_samples_per_second": 59.737,
      "eval_steps_per_second": 3.896,
      "step": 56
    },
    {
      "epoch": 16.86,
      "eval_accuracy": 0.6304347826086957,
      "eval_loss": 1.0714055299758911,
      "eval_runtime": 0.7702,
      "eval_samples_per_second": 59.726,
      "eval_steps_per_second": 3.895,
      "step": 59
    },
    {
      "epoch": 17.14,
      "learning_rate": 1.8421052631578944e-05,
      "loss": 1.0824,
      "step": 60
    },
    {
      "epoch": 18.0,
      "eval_accuracy": 0.6521739130434783,
      "eval_loss": 1.0453197956085205,
      "eval_runtime": 0.7803,
      "eval_samples_per_second": 58.955,
      "eval_steps_per_second": 3.845,
      "step": 63
    },
    {
      "epoch": 18.86,
      "eval_accuracy": 0.6521739130434783,
      "eval_loss": 1.0149754285812378,
      "eval_runtime": 0.7572,
      "eval_samples_per_second": 60.752,
      "eval_steps_per_second": 3.962,
      "step": 66
    },
    {
      "epoch": 20.0,
      "learning_rate": 1.5350877192982453e-05,
      "loss": 1.0535,
      "step": 70
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 0.6521739130434783,
      "eval_loss": 0.9924822449684143,
      "eval_runtime": 0.7866,
      "eval_samples_per_second": 58.483,
      "eval_steps_per_second": 3.814,
      "step": 70
    },
    {
      "epoch": 20.86,
      "eval_accuracy": 0.6304347826086957,
      "eval_loss": 0.9778493642807007,
      "eval_runtime": 0.8042,
      "eval_samples_per_second": 57.201,
      "eval_steps_per_second": 3.731,
      "step": 73
    },
    {
      "epoch": 22.0,
      "eval_accuracy": 0.6521739130434783,
      "eval_loss": 0.9570326805114746,
      "eval_runtime": 0.7982,
      "eval_samples_per_second": 57.628,
      "eval_steps_per_second": 3.758,
      "step": 77
    },
    {
      "epoch": 22.86,
      "learning_rate": 1.2280701754385963e-05,
      "loss": 0.994,
      "step": 80
    },
    {
      "epoch": 22.86,
      "eval_accuracy": 0.6304347826086957,
      "eval_loss": 0.9441110491752625,
      "eval_runtime": 0.7429,
      "eval_samples_per_second": 61.919,
      "eval_steps_per_second": 4.038,
      "step": 80
    },
    {
      "epoch": 24.0,
      "eval_accuracy": 0.6304347826086957,
      "eval_loss": 0.9246149659156799,
      "eval_runtime": 0.7415,
      "eval_samples_per_second": 62.033,
      "eval_steps_per_second": 4.046,
      "step": 84
    },
    {
      "epoch": 24.86,
      "eval_accuracy": 0.6304347826086957,
      "eval_loss": 0.9094782471656799,
      "eval_runtime": 0.755,
      "eval_samples_per_second": 60.928,
      "eval_steps_per_second": 3.974,
      "step": 87
    },
    {
      "epoch": 25.71,
      "learning_rate": 9.210526315789472e-06,
      "loss": 0.9554,
      "step": 90
    },
    {
      "epoch": 26.0,
      "eval_accuracy": 0.6304347826086957,
      "eval_loss": 0.8936623334884644,
      "eval_runtime": 0.7873,
      "eval_samples_per_second": 58.43,
      "eval_steps_per_second": 3.811,
      "step": 91
    },
    {
      "epoch": 26.86,
      "eval_accuracy": 0.6521739130434783,
      "eval_loss": 0.8924787044525146,
      "eval_runtime": 0.7504,
      "eval_samples_per_second": 61.301,
      "eval_steps_per_second": 3.998,
      "step": 94
    },
    {
      "epoch": 28.0,
      "eval_accuracy": 0.6521739130434783,
      "eval_loss": 0.8886207938194275,
      "eval_runtime": 0.748,
      "eval_samples_per_second": 61.494,
      "eval_steps_per_second": 4.01,
      "step": 98
    },
    {
      "epoch": 28.57,
      "learning_rate": 6.140350877192981e-06,
      "loss": 0.953,
      "step": 100
    },
    {
      "epoch": 28.86,
      "eval_accuracy": 0.6521739130434783,
      "eval_loss": 0.8804457187652588,
      "eval_runtime": 0.7633,
      "eval_samples_per_second": 60.262,
      "eval_steps_per_second": 3.93,
      "step": 101
    },
    {
      "epoch": 30.0,
      "eval_accuracy": 0.6521739130434783,
      "eval_loss": 0.8744495511054993,
      "eval_runtime": 0.76,
      "eval_samples_per_second": 60.524,
      "eval_steps_per_second": 3.947,
      "step": 105
    },
    {
      "epoch": 30.86,
      "eval_accuracy": 0.6521739130434783,
      "eval_loss": 0.875434398651123,
      "eval_runtime": 0.8019,
      "eval_samples_per_second": 57.364,
      "eval_steps_per_second": 3.741,
      "step": 108
    },
    {
      "epoch": 31.43,
      "learning_rate": 3.0701754385964907e-06,
      "loss": 0.9092,
      "step": 110
    },
    {
      "epoch": 32.0,
      "eval_accuracy": 0.6521739130434783,
      "eval_loss": 0.8732549548149109,
      "eval_runtime": 0.8029,
      "eval_samples_per_second": 57.294,
      "eval_steps_per_second": 3.737,
      "step": 112
    },
    {
      "epoch": 32.86,
      "eval_accuracy": 0.6521739130434783,
      "eval_loss": 0.8725218772888184,
      "eval_runtime": 0.8099,
      "eval_samples_per_second": 56.796,
      "eval_steps_per_second": 3.704,
      "step": 115
    },
    {
      "epoch": 34.0,
      "eval_accuracy": 0.6521739130434783,
      "eval_loss": 0.8721739649772644,
      "eval_runtime": 0.7302,
      "eval_samples_per_second": 62.996,
      "eval_steps_per_second": 4.108,
      "step": 119
    },
    {
      "epoch": 34.29,
      "learning_rate": 0.0,
      "loss": 0.9247,
      "step": 120
    },
    {
      "epoch": 34.29,
      "eval_accuracy": 0.6521739130434783,
      "eval_loss": 0.8720169067382812,
      "eval_runtime": 0.7717,
      "eval_samples_per_second": 59.608,
      "eval_steps_per_second": 3.887,
      "step": 120
    },
    {
      "epoch": 34.29,
      "step": 120,
      "total_flos": 2.3770905934823424e+17,
      "train_loss": 1.0817648808161417,
      "train_runtime": 199.6882,
      "train_samples_per_second": 42.667,
      "train_steps_per_second": 0.601
    }
  ],
  "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
}