{
  "best_metric": 0.8358637669265491,
  "best_model_checkpoint": "swin-base-patch4-window7-224-in22k-MM_NMM_Classification_base_V10/checkpoint-133",
  "epoch": 6.909090909090909,
  "eval_steps": 500,
  "global_step": 133,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.5194805194805194,
      "grad_norm": 3.6334409713745117,
      "learning_rate": 3.571428571428572e-05,
      "loss": 1.0502,
      "step": 10
    },
    {
      "epoch": 0.987012987012987,
      "step": 19,
      "train_accuracy": 0.7569835260411337,
      "train_auc_class_0": 0.9160739393649986,
      "train_auc_class_1": 0.9059040692814828,
      "train_auc_class_2": 0.8730912427450218,
      "train_auc_overall": 0.8983564171305011,
      "train_loss": 0.5773088932037354,
      "train_runtime": 897.4347,
      "train_samples_per_second": 10.89,
      "train_steps_per_second": 0.086
    },
    {
      "epoch": 0.987012987012987,
      "eval_accuracy": 0.7911366434140337,
      "eval_auc_class_0": 0.9403359734287176,
      "eval_auc_class_1": 0.9235862442916344,
      "eval_auc_class_2": 0.8963866320052807,
      "eval_auc_overall": 0.9201029499085442,
      "eval_loss": 0.5198014974594116,
      "eval_runtime": 265.4844,
      "eval_samples_per_second": 9.179,
      "eval_steps_per_second": 0.075,
      "step": 19
    },
    {
      "epoch": 1.0389610389610389,
      "grad_norm": 3.133422374725342,
      "learning_rate": 4.747899159663866e-05,
      "loss": 0.6929,
      "step": 20
    },
    {
      "epoch": 1.5584415584415585,
      "grad_norm": 3.1555869579315186,
      "learning_rate": 4.327731092436975e-05,
      "loss": 0.5585,
      "step": 30
    },
    {
      "epoch": 1.974025974025974,
      "step": 38,
      "train_accuracy": 0.8060984344622941,
      "train_auc_class_0": 0.9480363794091431,
      "train_auc_class_1": 0.9353479593951236,
      "train_auc_class_2": 0.9102701365313304,
      "train_auc_overall": 0.9312181584451991,
      "train_loss": 0.47007134556770325,
      "train_runtime": 906.6059,
      "train_samples_per_second": 10.78,
      "train_steps_per_second": 0.085
    },
    {
      "epoch": 1.974025974025974,
      "eval_accuracy": 0.8104226508001642,
      "eval_auc_class_0": 0.9510794583546245,
      "eval_auc_class_1": 0.9375873639586919,
      "eval_auc_class_2": 0.9143061067603285,
      "eval_auc_overall": 0.9343243096912149,
      "eval_loss": 0.4643743932247162,
      "eval_runtime": 266.4476,
      "eval_samples_per_second": 9.146,
      "eval_steps_per_second": 0.075,
      "step": 38
    },
    {
      "epoch": 2.0779220779220777,
      "grad_norm": 2.1339571475982666,
      "learning_rate": 3.907563025210084e-05,
      "loss": 0.5114,
      "step": 40
    },
    {
      "epoch": 2.5974025974025974,
      "grad_norm": 2.0463883876800537,
      "learning_rate": 3.487394957983193e-05,
      "loss": 0.47,
      "step": 50
    },
    {
      "epoch": 2.961038961038961,
      "step": 57,
      "train_accuracy": 0.819093420648726,
      "train_auc_class_0": 0.959281252652914,
      "train_auc_class_1": 0.9451989583224498,
      "train_auc_class_2": 0.9209859149104825,
      "train_auc_overall": 0.9418220419619487,
      "train_loss": 0.4309811294078827,
      "train_runtime": 984.0694,
      "train_samples_per_second": 9.931,
      "train_steps_per_second": 0.078
    },
    {
      "epoch": 2.961038961038961,
      "eval_accuracy": 0.8276569552728765,
      "eval_auc_class_0": 0.9621529552035428,
      "eval_auc_class_1": 0.9414093721062062,
      "eval_auc_class_2": 0.9226855511459174,
      "eval_auc_overall": 0.9420826261518888,
      "eval_loss": 0.4235168695449829,
      "eval_runtime": 284.7667,
      "eval_samples_per_second": 8.558,
      "eval_steps_per_second": 0.07,
      "step": 57
    },
    {
      "epoch": 3.116883116883117,
      "grad_norm": 2.500288963317871,
      "learning_rate": 3.0672268907563026e-05,
      "loss": 0.4485,
      "step": 60
    },
    {
      "epoch": 3.6363636363636362,
      "grad_norm": 2.1541225910186768,
      "learning_rate": 2.647058823529412e-05,
      "loss": 0.4487,
      "step": 70
    },
    {
      "epoch": 4.0,
      "step": 77,
      "train_accuracy": 0.8312698250281387,
      "train_auc_class_0": 0.9636362430493655,
      "train_auc_class_1": 0.9487531692630246,
      "train_auc_class_2": 0.928211332215209,
      "train_auc_overall": 0.9468669148425329,
      "train_loss": 0.41109779477119446,
      "train_runtime": 958.2523,
      "train_samples_per_second": 10.199,
      "train_steps_per_second": 0.08
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.826015592942142,
      "eval_auc_class_0": 0.9611033043774485,
      "eval_auc_class_1": 0.9434057595955114,
      "eval_auc_class_2": 0.9248328729019133,
      "eval_auc_overall": 0.9431139789582911,
      "eval_loss": 0.4228733479976654,
      "eval_runtime": 284.7186,
      "eval_samples_per_second": 8.559,
      "eval_steps_per_second": 0.07,
      "step": 77
    },
    {
      "epoch": 4.1558441558441555,
      "grad_norm": 2.5099682807922363,
      "learning_rate": 2.226890756302521e-05,
      "loss": 0.4217,
      "step": 80
    },
    {
      "epoch": 4.675324675324675,
      "grad_norm": 2.6273367404937744,
      "learning_rate": 1.8067226890756303e-05,
      "loss": 0.4076,
      "step": 90
    },
    {
      "epoch": 4.987012987012987,
      "step": 96,
      "train_accuracy": 0.8339302159009516,
      "train_auc_class_0": 0.9658112345600407,
      "train_auc_class_1": 0.9512605003604601,
      "train_auc_class_2": 0.931225975691603,
      "train_auc_overall": 0.9494325702040346,
      "train_loss": 0.4025966227054596,
      "train_runtime": 991.1294,
      "train_samples_per_second": 9.86,
      "train_steps_per_second": 0.078
    },
    {
      "epoch": 4.987012987012987,
      "eval_accuracy": 0.8297086581862946,
      "eval_auc_class_0": 0.9624020609776871,
      "eval_auc_class_1": 0.9453678176703179,
      "eval_auc_class_2": 0.9253326922606775,
      "eval_auc_overall": 0.9443675236362274,
      "eval_loss": 0.41654419898986816,
      "eval_runtime": 281.0044,
      "eval_samples_per_second": 8.672,
      "eval_steps_per_second": 0.071,
      "step": 96
    },
    {
      "epoch": 5.194805194805195,
      "grad_norm": 2.5981998443603516,
      "learning_rate": 1.3865546218487396e-05,
      "loss": 0.4229,
      "step": 100
    },
    {
      "epoch": 5.714285714285714,
      "grad_norm": 2.4040989875793457,
      "learning_rate": 9.663865546218488e-06,
      "loss": 0.4182,
      "step": 110
    },
    {
      "epoch": 5.974025974025974,
      "step": 115,
      "train_accuracy": 0.8407858385347385,
      "train_auc_class_0": 0.9691956895453967,
      "train_auc_class_1": 0.9539528502470981,
      "train_auc_class_2": 0.9353985798608017,
      "train_auc_overall": 0.9528490398844321,
      "train_loss": 0.3878542184829712,
      "train_runtime": 964.3964,
      "train_samples_per_second": 10.134,
      "train_steps_per_second": 0.08
    },
    {
      "epoch": 5.974025974025974,
      "eval_accuracy": 0.83258104226508,
      "eval_auc_class_0": 0.9633271589167093,
      "eval_auc_class_1": 0.9464734008654533,
      "eval_auc_class_2": 0.9275261929791678,
      "eval_auc_overall": 0.9457755842537768,
      "eval_loss": 0.4134509563446045,
      "eval_runtime": 281.0951,
      "eval_samples_per_second": 8.67,
      "eval_steps_per_second": 0.071,
      "step": 115
    },
    {
      "epoch": 6.233766233766234,
      "grad_norm": 1.9785808324813843,
      "learning_rate": 5.46218487394958e-06,
      "loss": 0.396,
      "step": 120
    },
    {
      "epoch": 6.753246753246753,
      "grad_norm": 1.799540400505066,
      "learning_rate": 1.2605042016806724e-06,
      "loss": 0.3874,
      "step": 130
    },
    {
      "epoch": 6.909090909090909,
      "step": 133,
      "train_accuracy": 0.8431392612299191,
      "train_auc_class_0": 0.9683630724988327,
      "train_auc_class_1": 0.9514347173174106,
      "train_auc_class_2": 0.9349089785503817,
      "train_auc_overall": 0.951568922788875,
      "train_loss": 0.39069029688835144,
      "train_runtime": 970.3934,
      "train_samples_per_second": 10.071,
      "train_steps_per_second": 0.079
    },
    {
      "epoch": 6.909090909090909,
      "eval_accuracy": 0.8358637669265491,
      "eval_auc_class_0": 0.9637412706523591,
      "eval_auc_class_1": 0.9464505145891209,
      "eval_auc_class_2": 0.928635846283816,
      "eval_auc_overall": 0.9462758771750988,
      "eval_loss": 0.4069300889968872,
      "eval_runtime": 282.2427,
      "eval_samples_per_second": 8.634,
      "eval_steps_per_second": 0.071,
      "step": 133
    },
    {
      "epoch": 6.909090909090909,
      "step": 133,
      "total_flos": 5.296065448002933e+18,
      "train_loss": 0.5080686431182059,
      "train_runtime": 15650.1833,
      "train_samples_per_second": 4.371,
      "train_steps_per_second": 0.008
    }
  ],
  "logging_steps": 10,
  "max_steps": 133,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "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": 5.296065448002933e+18,
  "train_batch_size": 128,
  "trial_name": null,
  "trial_params": null
}