{
  "best_metric": 0.7391304347826086,
  "best_model_checkpoint": "swinv2-tiny-patch4-window8-256-DMAE-da-colab/checkpoint-138",
  "epoch": 38.26086956521739,
  "eval_steps": 500,
  "global_step": 440,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.8695652173913043,
      "grad_norm": 16.909873962402344,
      "learning_rate": 9.090909090909091e-06,
      "loss": 1.3823,
      "step": 10
    },
    {
      "epoch": 0.9565217391304348,
      "eval_accuracy": 0.1956521739130435,
      "eval_loss": 1.4058117866516113,
      "eval_runtime": 0.9924,
      "eval_samples_per_second": 46.35,
      "eval_steps_per_second": 3.023,
      "step": 11
    },
    {
      "epoch": 1.7391304347826086,
      "grad_norm": 19.725204467773438,
      "learning_rate": 1.8181818181818182e-05,
      "loss": 1.3366,
      "step": 20
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.1956521739130435,
      "eval_loss": 1.4482489824295044,
      "eval_runtime": 0.7643,
      "eval_samples_per_second": 60.185,
      "eval_steps_per_second": 3.925,
      "step": 23
    },
    {
      "epoch": 2.608695652173913,
      "grad_norm": 25.69033432006836,
      "learning_rate": 2.7272727272727273e-05,
      "loss": 1.2352,
      "step": 30
    },
    {
      "epoch": 2.9565217391304346,
      "eval_accuracy": 0.45652173913043476,
      "eval_loss": 1.230912685394287,
      "eval_runtime": 0.8016,
      "eval_samples_per_second": 57.387,
      "eval_steps_per_second": 3.743,
      "step": 34
    },
    {
      "epoch": 3.4782608695652173,
      "grad_norm": 22.049612045288086,
      "learning_rate": 3.6363636363636364e-05,
      "loss": 1.1374,
      "step": 40
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6086956521739131,
      "eval_loss": 1.1031465530395508,
      "eval_runtime": 0.9885,
      "eval_samples_per_second": 46.536,
      "eval_steps_per_second": 3.035,
      "step": 46
    },
    {
      "epoch": 4.3478260869565215,
      "grad_norm": 29.715892791748047,
      "learning_rate": 3.93939393939394e-05,
      "loss": 1.0344,
      "step": 50
    },
    {
      "epoch": 4.956521739130435,
      "eval_accuracy": 0.5869565217391305,
      "eval_loss": 1.0230214595794678,
      "eval_runtime": 0.7523,
      "eval_samples_per_second": 61.145,
      "eval_steps_per_second": 3.988,
      "step": 57
    },
    {
      "epoch": 5.217391304347826,
      "grad_norm": 42.69334030151367,
      "learning_rate": 3.838383838383839e-05,
      "loss": 0.8772,
      "step": 60
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.6521739130434783,
      "eval_loss": 0.9114828705787659,
      "eval_runtime": 0.772,
      "eval_samples_per_second": 59.587,
      "eval_steps_per_second": 3.886,
      "step": 69
    },
    {
      "epoch": 6.086956521739131,
      "grad_norm": 43.39430618286133,
      "learning_rate": 3.7373737373737376e-05,
      "loss": 0.8105,
      "step": 70
    },
    {
      "epoch": 6.956521739130435,
      "grad_norm": 46.86561584472656,
      "learning_rate": 3.6363636363636364e-05,
      "loss": 0.7321,
      "step": 80
    },
    {
      "epoch": 6.956521739130435,
      "eval_accuracy": 0.6521739130434783,
      "eval_loss": 0.8857656717300415,
      "eval_runtime": 1.0277,
      "eval_samples_per_second": 44.76,
      "eval_steps_per_second": 2.919,
      "step": 80
    },
    {
      "epoch": 7.826086956521739,
      "grad_norm": 42.23720169067383,
      "learning_rate": 3.535353535353536e-05,
      "loss": 0.6319,
      "step": 90
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.6521739130434783,
      "eval_loss": 0.866457998752594,
      "eval_runtime": 0.7747,
      "eval_samples_per_second": 59.38,
      "eval_steps_per_second": 3.873,
      "step": 92
    },
    {
      "epoch": 8.695652173913043,
      "grad_norm": 42.64775466918945,
      "learning_rate": 3.434343434343435e-05,
      "loss": 0.6438,
      "step": 100
    },
    {
      "epoch": 8.956521739130435,
      "eval_accuracy": 0.717391304347826,
      "eval_loss": 0.7737635970115662,
      "eval_runtime": 0.7744,
      "eval_samples_per_second": 59.4,
      "eval_steps_per_second": 3.874,
      "step": 103
    },
    {
      "epoch": 9.565217391304348,
      "grad_norm": 60.89689636230469,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 0.4714,
      "step": 110
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.6304347826086957,
      "eval_loss": 0.8491604924201965,
      "eval_runtime": 1.0319,
      "eval_samples_per_second": 44.578,
      "eval_steps_per_second": 2.907,
      "step": 115
    },
    {
      "epoch": 10.434782608695652,
      "grad_norm": 47.14131546020508,
      "learning_rate": 3.232323232323232e-05,
      "loss": 0.433,
      "step": 120
    },
    {
      "epoch": 10.956521739130435,
      "eval_accuracy": 0.6956521739130435,
      "eval_loss": 0.8385751247406006,
      "eval_runtime": 0.7991,
      "eval_samples_per_second": 57.561,
      "eval_steps_per_second": 3.754,
      "step": 126
    },
    {
      "epoch": 11.304347826086957,
      "grad_norm": 40.993927001953125,
      "learning_rate": 3.131313131313132e-05,
      "loss": 0.4793,
      "step": 130
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.7391304347826086,
      "eval_loss": 0.939383327960968,
      "eval_runtime": 0.7851,
      "eval_samples_per_second": 58.594,
      "eval_steps_per_second": 3.821,
      "step": 138
    },
    {
      "epoch": 12.173913043478262,
      "grad_norm": 48.550899505615234,
      "learning_rate": 3.0303030303030306e-05,
      "loss": 0.4769,
      "step": 140
    },
    {
      "epoch": 12.956521739130435,
      "eval_accuracy": 0.6521739130434783,
      "eval_loss": 0.9470547437667847,
      "eval_runtime": 1.0593,
      "eval_samples_per_second": 43.423,
      "eval_steps_per_second": 2.832,
      "step": 149
    },
    {
      "epoch": 13.043478260869565,
      "grad_norm": 84.14537811279297,
      "learning_rate": 2.9292929292929297e-05,
      "loss": 0.4308,
      "step": 150
    },
    {
      "epoch": 13.91304347826087,
      "grad_norm": 54.65965270996094,
      "learning_rate": 2.8282828282828285e-05,
      "loss": 0.3872,
      "step": 160
    },
    {
      "epoch": 14.0,
      "eval_accuracy": 0.6086956521739131,
      "eval_loss": 1.1526373624801636,
      "eval_runtime": 0.7735,
      "eval_samples_per_second": 59.47,
      "eval_steps_per_second": 3.878,
      "step": 161
    },
    {
      "epoch": 14.782608695652174,
      "grad_norm": 63.24810791015625,
      "learning_rate": 2.7272727272727273e-05,
      "loss": 0.3906,
      "step": 170
    },
    {
      "epoch": 14.956521739130435,
      "eval_accuracy": 0.6521739130434783,
      "eval_loss": 1.057466983795166,
      "eval_runtime": 0.7776,
      "eval_samples_per_second": 59.156,
      "eval_steps_per_second": 3.858,
      "step": 172
    },
    {
      "epoch": 15.652173913043478,
      "grad_norm": 76.85520935058594,
      "learning_rate": 2.6262626262626265e-05,
      "loss": 0.3798,
      "step": 180
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.6956521739130435,
      "eval_loss": 1.0592772960662842,
      "eval_runtime": 0.8046,
      "eval_samples_per_second": 57.172,
      "eval_steps_per_second": 3.729,
      "step": 184
    },
    {
      "epoch": 16.52173913043478,
      "grad_norm": 43.57295608520508,
      "learning_rate": 2.5252525252525253e-05,
      "loss": 0.3377,
      "step": 190
    },
    {
      "epoch": 16.956521739130434,
      "eval_accuracy": 0.6086956521739131,
      "eval_loss": 1.0782535076141357,
      "eval_runtime": 0.7945,
      "eval_samples_per_second": 57.9,
      "eval_steps_per_second": 3.776,
      "step": 195
    },
    {
      "epoch": 17.391304347826086,
      "grad_norm": 55.173805236816406,
      "learning_rate": 2.4242424242424244e-05,
      "loss": 0.3919,
      "step": 200
    },
    {
      "epoch": 18.0,
      "eval_accuracy": 0.6521739130434783,
      "eval_loss": 1.1066713333129883,
      "eval_runtime": 0.7682,
      "eval_samples_per_second": 59.879,
      "eval_steps_per_second": 3.905,
      "step": 207
    },
    {
      "epoch": 18.26086956521739,
      "grad_norm": 44.900856018066406,
      "learning_rate": 2.3232323232323232e-05,
      "loss": 0.3631,
      "step": 210
    },
    {
      "epoch": 18.956521739130434,
      "eval_accuracy": 0.6739130434782609,
      "eval_loss": 1.1017966270446777,
      "eval_runtime": 0.7687,
      "eval_samples_per_second": 59.84,
      "eval_steps_per_second": 3.903,
      "step": 218
    },
    {
      "epoch": 19.130434782608695,
      "grad_norm": 45.883216857910156,
      "learning_rate": 2.2222222222222227e-05,
      "loss": 0.3433,
      "step": 220
    },
    {
      "epoch": 20.0,
      "grad_norm": 29.03622055053711,
      "learning_rate": 2.121212121212121e-05,
      "loss": 0.2762,
      "step": 230
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 0.6521739130434783,
      "eval_loss": 1.147935152053833,
      "eval_runtime": 1.0114,
      "eval_samples_per_second": 45.482,
      "eval_steps_per_second": 2.966,
      "step": 230
    },
    {
      "epoch": 20.869565217391305,
      "grad_norm": 49.36585235595703,
      "learning_rate": 2.0202020202020206e-05,
      "loss": 0.2935,
      "step": 240
    },
    {
      "epoch": 20.956521739130434,
      "eval_accuracy": 0.6956521739130435,
      "eval_loss": 1.1054612398147583,
      "eval_runtime": 0.7595,
      "eval_samples_per_second": 60.565,
      "eval_steps_per_second": 3.95,
      "step": 241
    },
    {
      "epoch": 21.73913043478261,
      "grad_norm": 34.57924270629883,
      "learning_rate": 1.9191919191919194e-05,
      "loss": 0.3029,
      "step": 250
    },
    {
      "epoch": 22.0,
      "eval_accuracy": 0.6739130434782609,
      "eval_loss": 1.120306372642517,
      "eval_runtime": 0.7563,
      "eval_samples_per_second": 60.82,
      "eval_steps_per_second": 3.967,
      "step": 253
    },
    {
      "epoch": 22.608695652173914,
      "grad_norm": 50.57252502441406,
      "learning_rate": 1.8181818181818182e-05,
      "loss": 0.2857,
      "step": 260
    },
    {
      "epoch": 22.956521739130434,
      "eval_accuracy": 0.6304347826086957,
      "eval_loss": 1.2819756269454956,
      "eval_runtime": 0.7606,
      "eval_samples_per_second": 60.476,
      "eval_steps_per_second": 3.944,
      "step": 264
    },
    {
      "epoch": 23.47826086956522,
      "grad_norm": 32.393211364746094,
      "learning_rate": 1.7171717171717173e-05,
      "loss": 0.2603,
      "step": 270
    },
    {
      "epoch": 24.0,
      "eval_accuracy": 0.6304347826086957,
      "eval_loss": 1.2550427913665771,
      "eval_runtime": 0.7542,
      "eval_samples_per_second": 60.991,
      "eval_steps_per_second": 3.978,
      "step": 276
    },
    {
      "epoch": 24.347826086956523,
      "grad_norm": 53.81830596923828,
      "learning_rate": 1.616161616161616e-05,
      "loss": 0.2162,
      "step": 280
    },
    {
      "epoch": 24.956521739130434,
      "eval_accuracy": 0.6739130434782609,
      "eval_loss": 1.1655433177947998,
      "eval_runtime": 0.7904,
      "eval_samples_per_second": 58.201,
      "eval_steps_per_second": 3.796,
      "step": 287
    },
    {
      "epoch": 25.217391304347824,
      "grad_norm": 61.60272216796875,
      "learning_rate": 1.5151515151515153e-05,
      "loss": 0.2465,
      "step": 290
    },
    {
      "epoch": 26.0,
      "eval_accuracy": 0.6739130434782609,
      "eval_loss": 1.2510898113250732,
      "eval_runtime": 0.796,
      "eval_samples_per_second": 57.789,
      "eval_steps_per_second": 3.769,
      "step": 299
    },
    {
      "epoch": 26.08695652173913,
      "grad_norm": 50.68645477294922,
      "learning_rate": 1.4141414141414143e-05,
      "loss": 0.2503,
      "step": 300
    },
    {
      "epoch": 26.956521739130434,
      "grad_norm": 62.43329620361328,
      "learning_rate": 1.3131313131313132e-05,
      "loss": 0.2238,
      "step": 310
    },
    {
      "epoch": 26.956521739130434,
      "eval_accuracy": 0.6304347826086957,
      "eval_loss": 1.3460850715637207,
      "eval_runtime": 1.0048,
      "eval_samples_per_second": 45.782,
      "eval_steps_per_second": 2.986,
      "step": 310
    },
    {
      "epoch": 27.82608695652174,
      "grad_norm": 56.05590057373047,
      "learning_rate": 1.2121212121212122e-05,
      "loss": 0.2271,
      "step": 320
    },
    {
      "epoch": 28.0,
      "eval_accuracy": 0.6304347826086957,
      "eval_loss": 1.3471860885620117,
      "eval_runtime": 0.7754,
      "eval_samples_per_second": 59.324,
      "eval_steps_per_second": 3.869,
      "step": 322
    },
    {
      "epoch": 28.695652173913043,
      "grad_norm": 36.40557098388672,
      "learning_rate": 1.1111111111111113e-05,
      "loss": 0.2694,
      "step": 330
    },
    {
      "epoch": 28.956521739130434,
      "eval_accuracy": 0.6304347826086957,
      "eval_loss": 1.4501123428344727,
      "eval_runtime": 0.746,
      "eval_samples_per_second": 61.663,
      "eval_steps_per_second": 4.022,
      "step": 333
    },
    {
      "epoch": 29.565217391304348,
      "grad_norm": 46.565956115722656,
      "learning_rate": 1.0101010101010103e-05,
      "loss": 0.1903,
      "step": 340
    },
    {
      "epoch": 30.0,
      "eval_accuracy": 0.6304347826086957,
      "eval_loss": 1.462881326675415,
      "eval_runtime": 1.0409,
      "eval_samples_per_second": 44.193,
      "eval_steps_per_second": 2.882,
      "step": 345
    },
    {
      "epoch": 30.434782608695652,
      "grad_norm": 92.46195220947266,
      "learning_rate": 9.090909090909091e-06,
      "loss": 0.2054,
      "step": 350
    },
    {
      "epoch": 30.956521739130434,
      "eval_accuracy": 0.6304347826086957,
      "eval_loss": 1.4671651124954224,
      "eval_runtime": 0.7747,
      "eval_samples_per_second": 59.374,
      "eval_steps_per_second": 3.872,
      "step": 356
    },
    {
      "epoch": 31.304347826086957,
      "grad_norm": 44.79200744628906,
      "learning_rate": 8.08080808080808e-06,
      "loss": 0.199,
      "step": 360
    },
    {
      "epoch": 32.0,
      "eval_accuracy": 0.6304347826086957,
      "eval_loss": 1.472485899925232,
      "eval_runtime": 0.7531,
      "eval_samples_per_second": 61.078,
      "eval_steps_per_second": 3.983,
      "step": 368
    },
    {
      "epoch": 32.17391304347826,
      "grad_norm": 34.2342643737793,
      "learning_rate": 7.070707070707071e-06,
      "loss": 0.2034,
      "step": 370
    },
    {
      "epoch": 32.95652173913044,
      "eval_accuracy": 0.6521739130434783,
      "eval_loss": 1.4506837129592896,
      "eval_runtime": 0.9808,
      "eval_samples_per_second": 46.899,
      "eval_steps_per_second": 3.059,
      "step": 379
    },
    {
      "epoch": 33.04347826086956,
      "grad_norm": 28.347145080566406,
      "learning_rate": 6.060606060606061e-06,
      "loss": 0.1959,
      "step": 380
    },
    {
      "epoch": 33.91304347826087,
      "grad_norm": 44.34228515625,
      "learning_rate": 5.0505050505050515e-06,
      "loss": 0.2048,
      "step": 390
    },
    {
      "epoch": 34.0,
      "eval_accuracy": 0.6304347826086957,
      "eval_loss": 1.4330270290374756,
      "eval_runtime": 0.9488,
      "eval_samples_per_second": 48.484,
      "eval_steps_per_second": 3.162,
      "step": 391
    },
    {
      "epoch": 34.78260869565217,
      "grad_norm": 38.90620803833008,
      "learning_rate": 4.04040404040404e-06,
      "loss": 0.1767,
      "step": 400
    },
    {
      "epoch": 34.95652173913044,
      "eval_accuracy": 0.6304347826086957,
      "eval_loss": 1.4638035297393799,
      "eval_runtime": 0.7554,
      "eval_samples_per_second": 60.894,
      "eval_steps_per_second": 3.971,
      "step": 402
    },
    {
      "epoch": 35.65217391304348,
      "grad_norm": 32.936920166015625,
      "learning_rate": 3.0303030303030305e-06,
      "loss": 0.1799,
      "step": 410
    },
    {
      "epoch": 36.0,
      "eval_accuracy": 0.6304347826086957,
      "eval_loss": 1.4231865406036377,
      "eval_runtime": 0.7731,
      "eval_samples_per_second": 59.501,
      "eval_steps_per_second": 3.88,
      "step": 414
    },
    {
      "epoch": 36.52173913043478,
      "grad_norm": 32.154483795166016,
      "learning_rate": 2.02020202020202e-06,
      "loss": 0.1903,
      "step": 420
    },
    {
      "epoch": 36.95652173913044,
      "eval_accuracy": 0.6304347826086957,
      "eval_loss": 1.450764536857605,
      "eval_runtime": 1.0455,
      "eval_samples_per_second": 43.997,
      "eval_steps_per_second": 2.869,
      "step": 425
    },
    {
      "epoch": 37.391304347826086,
      "grad_norm": 32.06462860107422,
      "learning_rate": 1.01010101010101e-06,
      "loss": 0.1864,
      "step": 430
    },
    {
      "epoch": 38.0,
      "eval_accuracy": 0.6304347826086957,
      "eval_loss": 1.4460225105285645,
      "eval_runtime": 0.78,
      "eval_samples_per_second": 58.972,
      "eval_steps_per_second": 3.846,
      "step": 437
    },
    {
      "epoch": 38.26086956521739,
      "grad_norm": 44.23124694824219,
      "learning_rate": 0.0,
      "loss": 0.1818,
      "step": 440
    },
    {
      "epoch": 38.26086956521739,
      "eval_accuracy": 0.6304347826086957,
      "eval_loss": 1.4456393718719482,
      "eval_runtime": 0.9819,
      "eval_samples_per_second": 46.85,
      "eval_steps_per_second": 3.055,
      "step": 440
    },
    {
      "epoch": 38.26086956521739,
      "step": 440,
      "total_flos": 8.989085534729011e+17,
      "train_loss": 0.4470971633087505,
      "train_runtime": 894.3938,
      "train_samples_per_second": 32.29,
      "train_steps_per_second": 0.492
    }
  ],
  "logging_steps": 10,
  "max_steps": 440,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 40,
  "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": 8.989085534729011e+17,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}