{
  "best_metric": 0.9262899262899262,
  "best_model_checkpoint": "leaftype-swin-tiny-patch4-window7-224-finetuned/checkpoint-140",
  "epoch": 28.235294117647058,
  "eval_steps": 500,
  "global_step": 360,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.78,
      "grad_norm": 2.9089090824127197,
      "learning_rate": 1.388888888888889e-05,
      "loss": 0.7327,
      "step": 10
    },
    {
      "epoch": 0.94,
      "eval_accuracy": 0.6044226044226044,
      "eval_loss": 0.615473210811615,
      "eval_runtime": 25.1718,
      "eval_samples_per_second": 16.169,
      "eval_steps_per_second": 0.516,
      "step": 12
    },
    {
      "epoch": 1.57,
      "grad_norm": 2.244661569595337,
      "learning_rate": 2.777777777777778e-05,
      "loss": 0.5479,
      "step": 20
    },
    {
      "epoch": 1.96,
      "eval_accuracy": 0.7936117936117936,
      "eval_loss": 0.4488780200481415,
      "eval_runtime": 20.5332,
      "eval_samples_per_second": 19.822,
      "eval_steps_per_second": 0.633,
      "step": 25
    },
    {
      "epoch": 2.35,
      "grad_norm": 3.1241555213928223,
      "learning_rate": 4.166666666666667e-05,
      "loss": 0.3205,
      "step": 30
    },
    {
      "epoch": 2.98,
      "eval_accuracy": 0.9115479115479116,
      "eval_loss": 0.2163781374692917,
      "eval_runtime": 20.8268,
      "eval_samples_per_second": 19.542,
      "eval_steps_per_second": 0.624,
      "step": 38
    },
    {
      "epoch": 3.14,
      "grad_norm": 3.5316970348358154,
      "learning_rate": 4.938271604938271e-05,
      "loss": 0.2492,
      "step": 40
    },
    {
      "epoch": 3.92,
      "grad_norm": 4.170050144195557,
      "learning_rate": 4.783950617283951e-05,
      "loss": 0.2386,
      "step": 50
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9115479115479116,
      "eval_loss": 0.21236275136470795,
      "eval_runtime": 20.7473,
      "eval_samples_per_second": 19.617,
      "eval_steps_per_second": 0.627,
      "step": 51
    },
    {
      "epoch": 4.71,
      "grad_norm": 2.763268232345581,
      "learning_rate": 4.62962962962963e-05,
      "loss": 0.1937,
      "step": 60
    },
    {
      "epoch": 4.94,
      "eval_accuracy": 0.9090909090909091,
      "eval_loss": 0.2348509132862091,
      "eval_runtime": 20.7257,
      "eval_samples_per_second": 19.637,
      "eval_steps_per_second": 0.627,
      "step": 63
    },
    {
      "epoch": 5.49,
      "grad_norm": 2.564243793487549,
      "learning_rate": 4.4753086419753084e-05,
      "loss": 0.1884,
      "step": 70
    },
    {
      "epoch": 5.96,
      "eval_accuracy": 0.9213759213759214,
      "eval_loss": 0.1752212643623352,
      "eval_runtime": 20.5939,
      "eval_samples_per_second": 19.763,
      "eval_steps_per_second": 0.631,
      "step": 76
    },
    {
      "epoch": 6.27,
      "grad_norm": 3.2644762992858887,
      "learning_rate": 4.3209876543209875e-05,
      "loss": 0.2186,
      "step": 80
    },
    {
      "epoch": 6.98,
      "eval_accuracy": 0.8697788697788698,
      "eval_loss": 0.3517865836620331,
      "eval_runtime": 20.5544,
      "eval_samples_per_second": 19.801,
      "eval_steps_per_second": 0.632,
      "step": 89
    },
    {
      "epoch": 7.06,
      "grad_norm": 4.700550079345703,
      "learning_rate": 4.166666666666667e-05,
      "loss": 0.1652,
      "step": 90
    },
    {
      "epoch": 7.84,
      "grad_norm": 3.568648099899292,
      "learning_rate": 4.012345679012346e-05,
      "loss": 0.1833,
      "step": 100
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9017199017199017,
      "eval_loss": 0.24429580569267273,
      "eval_runtime": 20.5265,
      "eval_samples_per_second": 19.828,
      "eval_steps_per_second": 0.633,
      "step": 102
    },
    {
      "epoch": 8.63,
      "grad_norm": 4.854302406311035,
      "learning_rate": 3.8580246913580246e-05,
      "loss": 0.1856,
      "step": 110
    },
    {
      "epoch": 8.94,
      "eval_accuracy": 0.9017199017199017,
      "eval_loss": 0.2491949051618576,
      "eval_runtime": 20.6467,
      "eval_samples_per_second": 19.713,
      "eval_steps_per_second": 0.63,
      "step": 114
    },
    {
      "epoch": 9.41,
      "grad_norm": 2.6214895248413086,
      "learning_rate": 3.7037037037037037e-05,
      "loss": 0.1605,
      "step": 120
    },
    {
      "epoch": 9.96,
      "eval_accuracy": 0.918918918918919,
      "eval_loss": 0.20045815408229828,
      "eval_runtime": 20.9241,
      "eval_samples_per_second": 19.451,
      "eval_steps_per_second": 0.621,
      "step": 127
    },
    {
      "epoch": 10.2,
      "grad_norm": 2.9705560207366943,
      "learning_rate": 3.5493827160493834e-05,
      "loss": 0.1631,
      "step": 130
    },
    {
      "epoch": 10.98,
      "grad_norm": 4.172762870788574,
      "learning_rate": 3.395061728395062e-05,
      "loss": 0.1934,
      "step": 140
    },
    {
      "epoch": 10.98,
      "eval_accuracy": 0.9262899262899262,
      "eval_loss": 0.17126818001270294,
      "eval_runtime": 20.7003,
      "eval_samples_per_second": 19.662,
      "eval_steps_per_second": 0.628,
      "step": 140
    },
    {
      "epoch": 11.76,
      "grad_norm": 2.098846912384033,
      "learning_rate": 3.240740740740741e-05,
      "loss": 0.186,
      "step": 150
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.9238329238329238,
      "eval_loss": 0.15729956328868866,
      "eval_runtime": 20.3679,
      "eval_samples_per_second": 19.982,
      "eval_steps_per_second": 0.638,
      "step": 153
    },
    {
      "epoch": 12.55,
      "grad_norm": 3.095176935195923,
      "learning_rate": 3.08641975308642e-05,
      "loss": 0.1407,
      "step": 160
    },
    {
      "epoch": 12.94,
      "eval_accuracy": 0.9213759213759214,
      "eval_loss": 0.16583235561847687,
      "eval_runtime": 21.1225,
      "eval_samples_per_second": 19.269,
      "eval_steps_per_second": 0.615,
      "step": 165
    },
    {
      "epoch": 13.33,
      "grad_norm": 5.465274333953857,
      "learning_rate": 2.9320987654320992e-05,
      "loss": 0.1655,
      "step": 170
    },
    {
      "epoch": 13.96,
      "eval_accuracy": 0.9213759213759214,
      "eval_loss": 0.15699075162410736,
      "eval_runtime": 20.6767,
      "eval_samples_per_second": 19.684,
      "eval_steps_per_second": 0.629,
      "step": 178
    },
    {
      "epoch": 14.12,
      "grad_norm": 3.078519105911255,
      "learning_rate": 2.777777777777778e-05,
      "loss": 0.1473,
      "step": 180
    },
    {
      "epoch": 14.9,
      "grad_norm": 2.30896258354187,
      "learning_rate": 2.623456790123457e-05,
      "loss": 0.1448,
      "step": 190
    },
    {
      "epoch": 14.98,
      "eval_accuracy": 0.9238329238329238,
      "eval_loss": 0.16221101582050323,
      "eval_runtime": 21.2348,
      "eval_samples_per_second": 19.167,
      "eval_steps_per_second": 0.612,
      "step": 191
    },
    {
      "epoch": 15.69,
      "grad_norm": 2.94296932220459,
      "learning_rate": 2.4691358024691357e-05,
      "loss": 0.1525,
      "step": 200
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.9164619164619164,
      "eval_loss": 0.21096211671829224,
      "eval_runtime": 21.3772,
      "eval_samples_per_second": 19.039,
      "eval_steps_per_second": 0.608,
      "step": 204
    },
    {
      "epoch": 16.47,
      "grad_norm": 4.256126403808594,
      "learning_rate": 2.314814814814815e-05,
      "loss": 0.1226,
      "step": 210
    },
    {
      "epoch": 16.94,
      "eval_accuracy": 0.9164619164619164,
      "eval_loss": 0.21088247001171112,
      "eval_runtime": 21.1755,
      "eval_samples_per_second": 19.22,
      "eval_steps_per_second": 0.614,
      "step": 216
    },
    {
      "epoch": 17.25,
      "grad_norm": 3.254096269607544,
      "learning_rate": 2.1604938271604937e-05,
      "loss": 0.1556,
      "step": 220
    },
    {
      "epoch": 17.96,
      "eval_accuracy": 0.9164619164619164,
      "eval_loss": 0.19139783084392548,
      "eval_runtime": 21.2277,
      "eval_samples_per_second": 19.173,
      "eval_steps_per_second": 0.612,
      "step": 229
    },
    {
      "epoch": 18.04,
      "grad_norm": 2.1944868564605713,
      "learning_rate": 2.006172839506173e-05,
      "loss": 0.1273,
      "step": 230
    },
    {
      "epoch": 18.82,
      "grad_norm": 3.8310389518737793,
      "learning_rate": 1.8518518518518518e-05,
      "loss": 0.1264,
      "step": 240
    },
    {
      "epoch": 18.98,
      "eval_accuracy": 0.9115479115479116,
      "eval_loss": 0.22510574758052826,
      "eval_runtime": 21.1643,
      "eval_samples_per_second": 19.231,
      "eval_steps_per_second": 0.614,
      "step": 242
    },
    {
      "epoch": 19.61,
      "grad_norm": 4.036985874176025,
      "learning_rate": 1.697530864197531e-05,
      "loss": 0.1249,
      "step": 250
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 0.8992628992628993,
      "eval_loss": 0.2862898111343384,
      "eval_runtime": 20.7633,
      "eval_samples_per_second": 19.602,
      "eval_steps_per_second": 0.626,
      "step": 255
    },
    {
      "epoch": 20.39,
      "grad_norm": 1.9880743026733398,
      "learning_rate": 1.54320987654321e-05,
      "loss": 0.1099,
      "step": 260
    },
    {
      "epoch": 20.94,
      "eval_accuracy": 0.9066339066339066,
      "eval_loss": 0.24470330774784088,
      "eval_runtime": 20.9944,
      "eval_samples_per_second": 19.386,
      "eval_steps_per_second": 0.619,
      "step": 267
    },
    {
      "epoch": 21.18,
      "grad_norm": 5.194756031036377,
      "learning_rate": 1.388888888888889e-05,
      "loss": 0.1398,
      "step": 270
    },
    {
      "epoch": 21.96,
      "grad_norm": 3.340437173843384,
      "learning_rate": 1.2345679012345678e-05,
      "loss": 0.1223,
      "step": 280
    },
    {
      "epoch": 21.96,
      "eval_accuracy": 0.9041769041769042,
      "eval_loss": 0.23575270175933838,
      "eval_runtime": 20.7211,
      "eval_samples_per_second": 19.642,
      "eval_steps_per_second": 0.627,
      "step": 280
    },
    {
      "epoch": 22.75,
      "grad_norm": 3.056072235107422,
      "learning_rate": 1.0802469135802469e-05,
      "loss": 0.1084,
      "step": 290
    },
    {
      "epoch": 22.98,
      "eval_accuracy": 0.9238329238329238,
      "eval_loss": 0.17134559154510498,
      "eval_runtime": 21.3775,
      "eval_samples_per_second": 19.039,
      "eval_steps_per_second": 0.608,
      "step": 293
    },
    {
      "epoch": 23.53,
      "grad_norm": 2.242640972137451,
      "learning_rate": 9.259259259259259e-06,
      "loss": 0.1054,
      "step": 300
    },
    {
      "epoch": 24.0,
      "eval_accuracy": 0.9115479115479116,
      "eval_loss": 0.20854736864566803,
      "eval_runtime": 20.9968,
      "eval_samples_per_second": 19.384,
      "eval_steps_per_second": 0.619,
      "step": 306
    },
    {
      "epoch": 24.31,
      "grad_norm": 4.247133731842041,
      "learning_rate": 7.71604938271605e-06,
      "loss": 0.1055,
      "step": 310
    },
    {
      "epoch": 24.94,
      "eval_accuracy": 0.9115479115479116,
      "eval_loss": 0.2001856565475464,
      "eval_runtime": 21.7847,
      "eval_samples_per_second": 18.683,
      "eval_steps_per_second": 0.597,
      "step": 318
    },
    {
      "epoch": 25.1,
      "grad_norm": 2.9881675243377686,
      "learning_rate": 6.172839506172839e-06,
      "loss": 0.1018,
      "step": 320
    },
    {
      "epoch": 25.88,
      "grad_norm": 2.0415101051330566,
      "learning_rate": 4.6296296296296296e-06,
      "loss": 0.1167,
      "step": 330
    },
    {
      "epoch": 25.96,
      "eval_accuracy": 0.914004914004914,
      "eval_loss": 0.22893595695495605,
      "eval_runtime": 21.0685,
      "eval_samples_per_second": 19.318,
      "eval_steps_per_second": 0.617,
      "step": 331
    },
    {
      "epoch": 26.67,
      "grad_norm": 4.408997535705566,
      "learning_rate": 3.0864197530864196e-06,
      "loss": 0.1156,
      "step": 340
    },
    {
      "epoch": 26.98,
      "eval_accuracy": 0.9115479115479116,
      "eval_loss": 0.18885986506938934,
      "eval_runtime": 20.3137,
      "eval_samples_per_second": 20.036,
      "eval_steps_per_second": 0.64,
      "step": 344
    },
    {
      "epoch": 27.45,
      "grad_norm": 3.3847527503967285,
      "learning_rate": 1.5432098765432098e-06,
      "loss": 0.0919,
      "step": 350
    },
    {
      "epoch": 28.0,
      "eval_accuracy": 0.9115479115479116,
      "eval_loss": 0.2014942616224289,
      "eval_runtime": 21.029,
      "eval_samples_per_second": 19.354,
      "eval_steps_per_second": 0.618,
      "step": 357
    },
    {
      "epoch": 28.24,
      "grad_norm": 3.328054666519165,
      "learning_rate": 0.0,
      "loss": 0.0827,
      "step": 360
    },
    {
      "epoch": 28.24,
      "eval_accuracy": 0.9090909090909091,
      "eval_loss": 0.202056884765625,
      "eval_runtime": 20.7948,
      "eval_samples_per_second": 19.572,
      "eval_steps_per_second": 0.625,
      "step": 360
    },
    {
      "epoch": 28.24,
      "step": 360,
      "total_flos": 1.1370123601689969e+18,
      "train_loss": 0.1815111721555392,
      "train_runtime": 2118.8582,
      "train_samples_per_second": 22.937,
      "train_steps_per_second": 0.17
    }
  ],
  "logging_steps": 10,
  "max_steps": 360,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 30,
  "save_steps": 500,
  "total_flos": 1.1370123601689969e+18,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}