{
  "best_metric": 0.6355140186915887,
  "best_model_checkpoint": "swinv2-tiny-patch4-window8-256-RH-6e-5\\checkpoint-312",
  "epoch": 40.0,
  "eval_steps": 500,
  "global_step": 320,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.411214953271028,
      "eval_loss": 4.618819713592529,
      "eval_runtime": 1.8874,
      "eval_samples_per_second": 56.691,
      "eval_steps_per_second": 3.709,
      "step": 8
    },
    {
      "epoch": 1.25,
      "learning_rate": 1.8750000000000002e-05,
      "loss": 4.5343,
      "step": 10
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.411214953271028,
      "eval_loss": 4.409636974334717,
      "eval_runtime": 1.8724,
      "eval_samples_per_second": 57.145,
      "eval_steps_per_second": 3.738,
      "step": 16
    },
    {
      "epoch": 2.5,
      "learning_rate": 3.7500000000000003e-05,
      "loss": 4.5396,
      "step": 20
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.411214953271028,
      "eval_loss": 3.6783952713012695,
      "eval_runtime": 1.8845,
      "eval_samples_per_second": 56.78,
      "eval_steps_per_second": 3.715,
      "step": 24
    },
    {
      "epoch": 3.75,
      "learning_rate": 5.625e-05,
      "loss": 3.621,
      "step": 30
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.411214953271028,
      "eval_loss": 2.479982852935791,
      "eval_runtime": 1.8659,
      "eval_samples_per_second": 57.344,
      "eval_steps_per_second": 3.751,
      "step": 32
    },
    {
      "epoch": 5.0,
      "learning_rate": 5.833333333333333e-05,
      "loss": 2.1758,
      "step": 40
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.411214953271028,
      "eval_loss": 1.2117878198623657,
      "eval_runtime": 1.864,
      "eval_samples_per_second": 57.405,
      "eval_steps_per_second": 3.755,
      "step": 40
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.5887850467289719,
      "eval_loss": 0.6790419816970825,
      "eval_runtime": 1.9149,
      "eval_samples_per_second": 55.876,
      "eval_steps_per_second": 3.655,
      "step": 48
    },
    {
      "epoch": 6.25,
      "learning_rate": 5.625e-05,
      "loss": 0.8871,
      "step": 50
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.5887850467289719,
      "eval_loss": 0.7903054356575012,
      "eval_runtime": 1.9304,
      "eval_samples_per_second": 55.428,
      "eval_steps_per_second": 3.626,
      "step": 56
    },
    {
      "epoch": 7.5,
      "learning_rate": 5.416666666666667e-05,
      "loss": 0.7484,
      "step": 60
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.5887850467289719,
      "eval_loss": 0.7640213370323181,
      "eval_runtime": 1.8494,
      "eval_samples_per_second": 57.856,
      "eval_steps_per_second": 3.785,
      "step": 64
    },
    {
      "epoch": 8.75,
      "learning_rate": 5.208333333333334e-05,
      "loss": 0.7414,
      "step": 70
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.5887850467289719,
      "eval_loss": 0.6789357662200928,
      "eval_runtime": 1.8705,
      "eval_samples_per_second": 57.204,
      "eval_steps_per_second": 3.742,
      "step": 72
    },
    {
      "epoch": 10.0,
      "learning_rate": 5e-05,
      "loss": 0.6868,
      "step": 80
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.5887850467289719,
      "eval_loss": 0.6770228743553162,
      "eval_runtime": 1.862,
      "eval_samples_per_second": 57.466,
      "eval_steps_per_second": 3.759,
      "step": 80
    },
    {
      "epoch": 11.0,
      "eval_accuracy": 0.5887850467289719,
      "eval_loss": 0.6774561405181885,
      "eval_runtime": 1.9195,
      "eval_samples_per_second": 55.745,
      "eval_steps_per_second": 3.647,
      "step": 88
    },
    {
      "epoch": 11.25,
      "learning_rate": 4.791666666666667e-05,
      "loss": 0.6771,
      "step": 90
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.5887850467289719,
      "eval_loss": 0.6992754340171814,
      "eval_runtime": 1.8744,
      "eval_samples_per_second": 57.084,
      "eval_steps_per_second": 3.734,
      "step": 96
    },
    {
      "epoch": 12.5,
      "learning_rate": 4.5833333333333334e-05,
      "loss": 0.7082,
      "step": 100
    },
    {
      "epoch": 13.0,
      "eval_accuracy": 0.5887850467289719,
      "eval_loss": 0.6765372157096863,
      "eval_runtime": 1.904,
      "eval_samples_per_second": 56.199,
      "eval_steps_per_second": 3.677,
      "step": 104
    },
    {
      "epoch": 13.75,
      "learning_rate": 4.375e-05,
      "loss": 0.6993,
      "step": 110
    },
    {
      "epoch": 14.0,
      "eval_accuracy": 0.5887850467289719,
      "eval_loss": 0.6745595335960388,
      "eval_runtime": 1.8884,
      "eval_samples_per_second": 56.661,
      "eval_steps_per_second": 3.707,
      "step": 112
    },
    {
      "epoch": 15.0,
      "learning_rate": 4.1666666666666665e-05,
      "loss": 0.6798,
      "step": 120
    },
    {
      "epoch": 15.0,
      "eval_accuracy": 0.5887850467289719,
      "eval_loss": 0.6758854985237122,
      "eval_runtime": 1.922,
      "eval_samples_per_second": 55.672,
      "eval_steps_per_second": 3.642,
      "step": 120
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.5887850467289719,
      "eval_loss": 0.6734067797660828,
      "eval_runtime": 1.8769,
      "eval_samples_per_second": 57.008,
      "eval_steps_per_second": 3.729,
      "step": 128
    },
    {
      "epoch": 16.25,
      "learning_rate": 3.958333333333333e-05,
      "loss": 0.6734,
      "step": 130
    },
    {
      "epoch": 17.0,
      "eval_accuracy": 0.5887850467289719,
      "eval_loss": 0.673911988735199,
      "eval_runtime": 1.9047,
      "eval_samples_per_second": 56.178,
      "eval_steps_per_second": 3.675,
      "step": 136
    },
    {
      "epoch": 17.5,
      "learning_rate": 3.7500000000000003e-05,
      "loss": 0.6832,
      "step": 140
    },
    {
      "epoch": 18.0,
      "eval_accuracy": 0.5887850467289719,
      "eval_loss": 0.7038590908050537,
      "eval_runtime": 1.9054,
      "eval_samples_per_second": 56.155,
      "eval_steps_per_second": 3.674,
      "step": 144
    },
    {
      "epoch": 18.75,
      "learning_rate": 3.541666666666667e-05,
      "loss": 0.6825,
      "step": 150
    },
    {
      "epoch": 19.0,
      "eval_accuracy": 0.5887850467289719,
      "eval_loss": 0.676673412322998,
      "eval_runtime": 1.975,
      "eval_samples_per_second": 54.178,
      "eval_steps_per_second": 3.544,
      "step": 152
    },
    {
      "epoch": 20.0,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 0.6663,
      "step": 160
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 0.5887850467289719,
      "eval_loss": 0.670670747756958,
      "eval_runtime": 1.8789,
      "eval_samples_per_second": 56.947,
      "eval_steps_per_second": 3.725,
      "step": 160
    },
    {
      "epoch": 21.0,
      "eval_accuracy": 0.5887850467289719,
      "eval_loss": 0.6798433661460876,
      "eval_runtime": 1.888,
      "eval_samples_per_second": 56.675,
      "eval_steps_per_second": 3.708,
      "step": 168
    },
    {
      "epoch": 21.25,
      "learning_rate": 3.125e-05,
      "loss": 0.6646,
      "step": 170
    },
    {
      "epoch": 22.0,
      "eval_accuracy": 0.5794392523364486,
      "eval_loss": 0.672282874584198,
      "eval_runtime": 1.8865,
      "eval_samples_per_second": 56.72,
      "eval_steps_per_second": 3.711,
      "step": 176
    },
    {
      "epoch": 22.5,
      "learning_rate": 2.9166666666666666e-05,
      "loss": 0.6764,
      "step": 180
    },
    {
      "epoch": 23.0,
      "eval_accuracy": 0.5887850467289719,
      "eval_loss": 0.6889127492904663,
      "eval_runtime": 1.8825,
      "eval_samples_per_second": 56.839,
      "eval_steps_per_second": 3.718,
      "step": 184
    },
    {
      "epoch": 23.75,
      "learning_rate": 2.7083333333333335e-05,
      "loss": 0.6808,
      "step": 190
    },
    {
      "epoch": 24.0,
      "eval_accuracy": 0.5887850467289719,
      "eval_loss": 0.6993632316589355,
      "eval_runtime": 2.065,
      "eval_samples_per_second": 51.816,
      "eval_steps_per_second": 3.39,
      "step": 192
    },
    {
      "epoch": 25.0,
      "learning_rate": 2.5e-05,
      "loss": 0.6766,
      "step": 200
    },
    {
      "epoch": 25.0,
      "eval_accuracy": 0.5887850467289719,
      "eval_loss": 0.6691157221794128,
      "eval_runtime": 1.8589,
      "eval_samples_per_second": 57.56,
      "eval_steps_per_second": 3.766,
      "step": 200
    },
    {
      "epoch": 26.0,
      "eval_accuracy": 0.5887850467289719,
      "eval_loss": 0.683734655380249,
      "eval_runtime": 1.8344,
      "eval_samples_per_second": 58.329,
      "eval_steps_per_second": 3.816,
      "step": 208
    },
    {
      "epoch": 26.25,
      "learning_rate": 2.2916666666666667e-05,
      "loss": 0.6698,
      "step": 210
    },
    {
      "epoch": 27.0,
      "eval_accuracy": 0.5700934579439252,
      "eval_loss": 0.673797607421875,
      "eval_runtime": 1.9525,
      "eval_samples_per_second": 54.802,
      "eval_steps_per_second": 3.585,
      "step": 216
    },
    {
      "epoch": 27.5,
      "learning_rate": 2.0833333333333333e-05,
      "loss": 0.6549,
      "step": 220
    },
    {
      "epoch": 28.0,
      "eval_accuracy": 0.5794392523364486,
      "eval_loss": 0.6694583892822266,
      "eval_runtime": 2.1597,
      "eval_samples_per_second": 49.545,
      "eval_steps_per_second": 3.241,
      "step": 224
    },
    {
      "epoch": 28.75,
      "learning_rate": 1.8750000000000002e-05,
      "loss": 0.6442,
      "step": 230
    },
    {
      "epoch": 29.0,
      "eval_accuracy": 0.5794392523364486,
      "eval_loss": 0.7157074213027954,
      "eval_runtime": 1.8589,
      "eval_samples_per_second": 57.56,
      "eval_steps_per_second": 3.766,
      "step": 232
    },
    {
      "epoch": 30.0,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 0.649,
      "step": 240
    },
    {
      "epoch": 30.0,
      "eval_accuracy": 0.6074766355140186,
      "eval_loss": 0.6725906133651733,
      "eval_runtime": 1.8929,
      "eval_samples_per_second": 56.526,
      "eval_steps_per_second": 3.698,
      "step": 240
    },
    {
      "epoch": 31.0,
      "eval_accuracy": 0.5794392523364486,
      "eval_loss": 0.6838603019714355,
      "eval_runtime": 1.8925,
      "eval_samples_per_second": 56.54,
      "eval_steps_per_second": 3.699,
      "step": 248
    },
    {
      "epoch": 31.25,
      "learning_rate": 1.4583333333333333e-05,
      "loss": 0.6388,
      "step": 250
    },
    {
      "epoch": 32.0,
      "eval_accuracy": 0.5887850467289719,
      "eval_loss": 0.679723858833313,
      "eval_runtime": 1.858,
      "eval_samples_per_second": 57.59,
      "eval_steps_per_second": 3.768,
      "step": 256
    },
    {
      "epoch": 32.5,
      "learning_rate": 1.25e-05,
      "loss": 0.6416,
      "step": 260
    },
    {
      "epoch": 33.0,
      "eval_accuracy": 0.5981308411214953,
      "eval_loss": 0.6713680624961853,
      "eval_runtime": 1.9354,
      "eval_samples_per_second": 55.284,
      "eval_steps_per_second": 3.617,
      "step": 264
    },
    {
      "epoch": 33.75,
      "learning_rate": 1.0416666666666666e-05,
      "loss": 0.6398,
      "step": 270
    },
    {
      "epoch": 34.0,
      "eval_accuracy": 0.6074766355140186,
      "eval_loss": 0.6730300188064575,
      "eval_runtime": 1.961,
      "eval_samples_per_second": 54.565,
      "eval_steps_per_second": 3.57,
      "step": 272
    },
    {
      "epoch": 35.0,
      "learning_rate": 8.333333333333334e-06,
      "loss": 0.6522,
      "step": 280
    },
    {
      "epoch": 35.0,
      "eval_accuracy": 0.5794392523364486,
      "eval_loss": 0.6953481435775757,
      "eval_runtime": 1.901,
      "eval_samples_per_second": 56.288,
      "eval_steps_per_second": 3.682,
      "step": 280
    },
    {
      "epoch": 36.0,
      "eval_accuracy": 0.5700934579439252,
      "eval_loss": 0.6608957052230835,
      "eval_runtime": 1.8975,
      "eval_samples_per_second": 56.391,
      "eval_steps_per_second": 3.689,
      "step": 288
    },
    {
      "epoch": 36.25,
      "learning_rate": 6.25e-06,
      "loss": 0.6376,
      "step": 290
    },
    {
      "epoch": 37.0,
      "eval_accuracy": 0.5794392523364486,
      "eval_loss": 0.6618574857711792,
      "eval_runtime": 1.9264,
      "eval_samples_per_second": 55.543,
      "eval_steps_per_second": 3.634,
      "step": 296
    },
    {
      "epoch": 37.5,
      "learning_rate": 4.166666666666667e-06,
      "loss": 0.6441,
      "step": 300
    },
    {
      "epoch": 38.0,
      "eval_accuracy": 0.6261682242990654,
      "eval_loss": 0.6653915047645569,
      "eval_runtime": 1.921,
      "eval_samples_per_second": 55.701,
      "eval_steps_per_second": 3.644,
      "step": 304
    },
    {
      "epoch": 38.75,
      "learning_rate": 2.0833333333333334e-06,
      "loss": 0.6149,
      "step": 310
    },
    {
      "epoch": 39.0,
      "eval_accuracy": 0.6355140186915887,
      "eval_loss": 0.666593611240387,
      "eval_runtime": 1.9335,
      "eval_samples_per_second": 55.341,
      "eval_steps_per_second": 3.62,
      "step": 312
    },
    {
      "epoch": 40.0,
      "learning_rate": 0.0,
      "loss": 0.623,
      "step": 320
    },
    {
      "epoch": 40.0,
      "eval_accuracy": 0.6355140186915887,
      "eval_loss": 0.6678972840309143,
      "eval_runtime": 1.8599,
      "eval_samples_per_second": 57.529,
      "eval_steps_per_second": 3.764,
      "step": 320
    },
    {
      "epoch": 40.0,
      "step": 320,
      "total_flos": 6.480838238876467e+17,
      "train_loss": 1.0566354274749756,
      "train_runtime": 583.642,
      "train_samples_per_second": 34.131,
      "train_steps_per_second": 0.548
    }
  ],
  "logging_steps": 10,
  "max_steps": 320,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 40,
  "save_steps": 500,
  "total_flos": 6.480838238876467e+17,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}