{
  "best_metric": 0.7692307692307693,
  "best_model_checkpoint": "swinv2-tiny-patch4-window8-256-dmae-humeda-DA/checkpoint-57",
  "epoch": 33.68421052631579,
  "eval_steps": 500,
  "global_step": 160,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.8421052631578947,
      "eval_accuracy": 0.1346153846153846,
      "eval_loss": 1.6341646909713745,
      "eval_runtime": 0.9419,
      "eval_samples_per_second": 55.206,
      "eval_steps_per_second": 2.123,
      "step": 4
    },
    {
      "epoch": 1.8947368421052633,
      "eval_accuracy": 0.2692307692307692,
      "eval_loss": 1.5368279218673706,
      "eval_runtime": 1.3102,
      "eval_samples_per_second": 39.688,
      "eval_steps_per_second": 1.526,
      "step": 9
    },
    {
      "epoch": 2.1052631578947367,
      "grad_norm": 4.0940937995910645,
      "learning_rate": 3.125e-05,
      "loss": 1.5734,
      "step": 10
    },
    {
      "epoch": 2.9473684210526314,
      "eval_accuracy": 0.4807692307692308,
      "eval_loss": 1.3871084451675415,
      "eval_runtime": 1.211,
      "eval_samples_per_second": 42.939,
      "eval_steps_per_second": 1.651,
      "step": 14
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.5192307692307693,
      "eval_loss": 1.2443485260009766,
      "eval_runtime": 0.9558,
      "eval_samples_per_second": 54.405,
      "eval_steps_per_second": 2.092,
      "step": 19
    },
    {
      "epoch": 4.2105263157894735,
      "grad_norm": 5.352275848388672,
      "learning_rate": 4.8611111111111115e-05,
      "loss": 1.3253,
      "step": 20
    },
    {
      "epoch": 4.842105263157895,
      "eval_accuracy": 0.5576923076923077,
      "eval_loss": 1.1184567213058472,
      "eval_runtime": 0.9411,
      "eval_samples_per_second": 55.256,
      "eval_steps_per_second": 2.125,
      "step": 23
    },
    {
      "epoch": 5.894736842105263,
      "eval_accuracy": 0.5961538461538461,
      "eval_loss": 1.0374407768249512,
      "eval_runtime": 0.9539,
      "eval_samples_per_second": 54.513,
      "eval_steps_per_second": 2.097,
      "step": 28
    },
    {
      "epoch": 6.315789473684211,
      "grad_norm": 8.018265724182129,
      "learning_rate": 4.5138888888888894e-05,
      "loss": 1.044,
      "step": 30
    },
    {
      "epoch": 6.947368421052632,
      "eval_accuracy": 0.6346153846153846,
      "eval_loss": 0.9798253178596497,
      "eval_runtime": 1.2838,
      "eval_samples_per_second": 40.504,
      "eval_steps_per_second": 1.558,
      "step": 33
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.6730769230769231,
      "eval_loss": 0.9701823592185974,
      "eval_runtime": 0.9841,
      "eval_samples_per_second": 52.842,
      "eval_steps_per_second": 2.032,
      "step": 38
    },
    {
      "epoch": 8.421052631578947,
      "grad_norm": 7.236335754394531,
      "learning_rate": 4.166666666666667e-05,
      "loss": 0.8487,
      "step": 40
    },
    {
      "epoch": 8.842105263157894,
      "eval_accuracy": 0.6538461538461539,
      "eval_loss": 0.9566704034805298,
      "eval_runtime": 0.9267,
      "eval_samples_per_second": 56.11,
      "eval_steps_per_second": 2.158,
      "step": 42
    },
    {
      "epoch": 9.894736842105264,
      "eval_accuracy": 0.6538461538461539,
      "eval_loss": 0.9291679859161377,
      "eval_runtime": 0.9628,
      "eval_samples_per_second": 54.006,
      "eval_steps_per_second": 2.077,
      "step": 47
    },
    {
      "epoch": 10.526315789473685,
      "grad_norm": 12.212766647338867,
      "learning_rate": 3.8194444444444444e-05,
      "loss": 0.7359,
      "step": 50
    },
    {
      "epoch": 10.947368421052632,
      "eval_accuracy": 0.6923076923076923,
      "eval_loss": 0.9042272567749023,
      "eval_runtime": 1.3007,
      "eval_samples_per_second": 39.978,
      "eval_steps_per_second": 1.538,
      "step": 52
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.7692307692307693,
      "eval_loss": 0.903170645236969,
      "eval_runtime": 0.9527,
      "eval_samples_per_second": 54.582,
      "eval_steps_per_second": 2.099,
      "step": 57
    },
    {
      "epoch": 12.631578947368421,
      "grad_norm": 16.251426696777344,
      "learning_rate": 3.472222222222222e-05,
      "loss": 0.6592,
      "step": 60
    },
    {
      "epoch": 12.842105263157894,
      "eval_accuracy": 0.6538461538461539,
      "eval_loss": 0.9060092568397522,
      "eval_runtime": 0.9363,
      "eval_samples_per_second": 55.539,
      "eval_steps_per_second": 2.136,
      "step": 61
    },
    {
      "epoch": 13.894736842105264,
      "eval_accuracy": 0.6538461538461539,
      "eval_loss": 0.9207931756973267,
      "eval_runtime": 0.9443,
      "eval_samples_per_second": 55.07,
      "eval_steps_per_second": 2.118,
      "step": 66
    },
    {
      "epoch": 14.736842105263158,
      "grad_norm": 11.413750648498535,
      "learning_rate": 3.125e-05,
      "loss": 0.6257,
      "step": 70
    },
    {
      "epoch": 14.947368421052632,
      "eval_accuracy": 0.6923076923076923,
      "eval_loss": 0.9272310733795166,
      "eval_runtime": 0.95,
      "eval_samples_per_second": 54.735,
      "eval_steps_per_second": 2.105,
      "step": 71
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.6730769230769231,
      "eval_loss": 1.0044364929199219,
      "eval_runtime": 1.3315,
      "eval_samples_per_second": 39.054,
      "eval_steps_per_second": 1.502,
      "step": 76
    },
    {
      "epoch": 16.842105263157894,
      "grad_norm": 8.319292068481445,
      "learning_rate": 2.777777777777778e-05,
      "loss": 0.5927,
      "step": 80
    },
    {
      "epoch": 16.842105263157894,
      "eval_accuracy": 0.7307692307692307,
      "eval_loss": 0.9176420569419861,
      "eval_runtime": 0.9566,
      "eval_samples_per_second": 54.36,
      "eval_steps_per_second": 2.091,
      "step": 80
    },
    {
      "epoch": 17.894736842105264,
      "eval_accuracy": 0.75,
      "eval_loss": 0.9261331558227539,
      "eval_runtime": 0.9451,
      "eval_samples_per_second": 55.02,
      "eval_steps_per_second": 2.116,
      "step": 85
    },
    {
      "epoch": 18.94736842105263,
      "grad_norm": 9.767424583435059,
      "learning_rate": 2.4305555555555558e-05,
      "loss": 0.5255,
      "step": 90
    },
    {
      "epoch": 18.94736842105263,
      "eval_accuracy": 0.6538461538461539,
      "eval_loss": 0.9058236479759216,
      "eval_runtime": 0.9515,
      "eval_samples_per_second": 54.649,
      "eval_steps_per_second": 2.102,
      "step": 90
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 0.75,
      "eval_loss": 0.933762788772583,
      "eval_runtime": 0.9497,
      "eval_samples_per_second": 54.756,
      "eval_steps_per_second": 2.106,
      "step": 95
    },
    {
      "epoch": 20.842105263157894,
      "eval_accuracy": 0.6923076923076923,
      "eval_loss": 0.9102750420570374,
      "eval_runtime": 1.4808,
      "eval_samples_per_second": 35.117,
      "eval_steps_per_second": 1.351,
      "step": 99
    },
    {
      "epoch": 21.05263157894737,
      "grad_norm": 9.32693862915039,
      "learning_rate": 2.0833333333333336e-05,
      "loss": 0.5098,
      "step": 100
    },
    {
      "epoch": 21.894736842105264,
      "eval_accuracy": 0.75,
      "eval_loss": 0.9329497218132019,
      "eval_runtime": 1.3581,
      "eval_samples_per_second": 38.289,
      "eval_steps_per_second": 1.473,
      "step": 104
    },
    {
      "epoch": 22.94736842105263,
      "eval_accuracy": 0.75,
      "eval_loss": 0.9886109828948975,
      "eval_runtime": 0.9789,
      "eval_samples_per_second": 53.123,
      "eval_steps_per_second": 2.043,
      "step": 109
    },
    {
      "epoch": 23.157894736842106,
      "grad_norm": 11.134294509887695,
      "learning_rate": 1.736111111111111e-05,
      "loss": 0.4347,
      "step": 110
    },
    {
      "epoch": 24.0,
      "eval_accuracy": 0.7307692307692307,
      "eval_loss": 0.9331239461898804,
      "eval_runtime": 0.9622,
      "eval_samples_per_second": 54.044,
      "eval_steps_per_second": 2.079,
      "step": 114
    },
    {
      "epoch": 24.842105263157894,
      "eval_accuracy": 0.6923076923076923,
      "eval_loss": 1.0085707902908325,
      "eval_runtime": 1.1355,
      "eval_samples_per_second": 45.796,
      "eval_steps_per_second": 1.761,
      "step": 118
    },
    {
      "epoch": 25.263157894736842,
      "grad_norm": 8.963732719421387,
      "learning_rate": 1.388888888888889e-05,
      "loss": 0.4269,
      "step": 120
    },
    {
      "epoch": 25.894736842105264,
      "eval_accuracy": 0.7307692307692307,
      "eval_loss": 1.0184217691421509,
      "eval_runtime": 1.4779,
      "eval_samples_per_second": 35.184,
      "eval_steps_per_second": 1.353,
      "step": 123
    },
    {
      "epoch": 26.94736842105263,
      "eval_accuracy": 0.7115384615384616,
      "eval_loss": 0.9697605967521667,
      "eval_runtime": 1.1941,
      "eval_samples_per_second": 43.549,
      "eval_steps_per_second": 1.675,
      "step": 128
    },
    {
      "epoch": 27.36842105263158,
      "grad_norm": 6.851988792419434,
      "learning_rate": 1.0416666666666668e-05,
      "loss": 0.42,
      "step": 130
    },
    {
      "epoch": 28.0,
      "eval_accuracy": 0.6923076923076923,
      "eval_loss": 0.987307608127594,
      "eval_runtime": 0.9659,
      "eval_samples_per_second": 53.835,
      "eval_steps_per_second": 2.071,
      "step": 133
    },
    {
      "epoch": 28.842105263157894,
      "eval_accuracy": 0.6923076923076923,
      "eval_loss": 0.9995985627174377,
      "eval_runtime": 1.0689,
      "eval_samples_per_second": 48.647,
      "eval_steps_per_second": 1.871,
      "step": 137
    },
    {
      "epoch": 29.473684210526315,
      "grad_norm": 8.500880241394043,
      "learning_rate": 6.944444444444445e-06,
      "loss": 0.4239,
      "step": 140
    },
    {
      "epoch": 29.894736842105264,
      "eval_accuracy": 0.6923076923076923,
      "eval_loss": 0.9857708215713501,
      "eval_runtime": 1.1807,
      "eval_samples_per_second": 44.041,
      "eval_steps_per_second": 1.694,
      "step": 142
    },
    {
      "epoch": 30.94736842105263,
      "eval_accuracy": 0.7307692307692307,
      "eval_loss": 0.9963864684104919,
      "eval_runtime": 1.32,
      "eval_samples_per_second": 39.393,
      "eval_steps_per_second": 1.515,
      "step": 147
    },
    {
      "epoch": 31.57894736842105,
      "grad_norm": 7.70914888381958,
      "learning_rate": 3.4722222222222224e-06,
      "loss": 0.3713,
      "step": 150
    },
    {
      "epoch": 32.0,
      "eval_accuracy": 0.7115384615384616,
      "eval_loss": 1.0337761640548706,
      "eval_runtime": 0.9818,
      "eval_samples_per_second": 52.966,
      "eval_steps_per_second": 2.037,
      "step": 152
    },
    {
      "epoch": 32.8421052631579,
      "eval_accuracy": 0.6923076923076923,
      "eval_loss": 1.047446370124817,
      "eval_runtime": 0.9575,
      "eval_samples_per_second": 54.308,
      "eval_steps_per_second": 2.089,
      "step": 156
    },
    {
      "epoch": 33.68421052631579,
      "grad_norm": 6.526296615600586,
      "learning_rate": 0.0,
      "loss": 0.4133,
      "step": 160
    },
    {
      "epoch": 33.68421052631579,
      "eval_accuracy": 0.6923076923076923,
      "eval_loss": 1.0444227457046509,
      "eval_runtime": 0.9985,
      "eval_samples_per_second": 52.081,
      "eval_steps_per_second": 2.003,
      "step": 160
    },
    {
      "epoch": 33.68421052631579,
      "step": 160,
      "total_flos": 6.642113378089697e+17,
      "train_loss": 0.6831471905112266,
      "train_runtime": 556.2019,
      "train_samples_per_second": 43.581,
      "train_steps_per_second": 0.288
    }
  ],
  "logging_steps": 10,
  "max_steps": 160,
  "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": 6.642113378089697e+17,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}