{
  "best_metric": 0.8071570576540755,
  "best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-eurosat/checkpoint-101",
  "epoch": 53.333333333333336,
  "eval_steps": 500,
  "global_step": 360,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.8888888888888888,
      "eval_accuracy": 0.6898608349900597,
      "eval_loss": 0.6376240849494934,
      "eval_runtime": 2.8661,
      "eval_samples_per_second": 175.501,
      "eval_steps_per_second": 5.583,
      "step": 6
    },
    {
      "epoch": 1.4814814814814814,
      "grad_norm": 9.892236709594727,
      "learning_rate": 1.388888888888889e-05,
      "loss": 0.6757,
      "step": 10
    },
    {
      "epoch": 1.925925925925926,
      "eval_accuracy": 0.6938369781312127,
      "eval_loss": 0.6052560806274414,
      "eval_runtime": 2.8417,
      "eval_samples_per_second": 177.006,
      "eval_steps_per_second": 5.63,
      "step": 13
    },
    {
      "epoch": 2.962962962962963,
      "grad_norm": 8.984474182128906,
      "learning_rate": 2.777777777777778e-05,
      "loss": 0.5472,
      "step": 20
    },
    {
      "epoch": 2.962962962962963,
      "eval_accuracy": 0.7256461232604374,
      "eval_loss": 0.5903410315513611,
      "eval_runtime": 2.7461,
      "eval_samples_per_second": 183.169,
      "eval_steps_per_second": 5.826,
      "step": 20
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.731610337972167,
      "eval_loss": 0.5782421827316284,
      "eval_runtime": 2.7061,
      "eval_samples_per_second": 185.878,
      "eval_steps_per_second": 5.913,
      "step": 27
    },
    {
      "epoch": 4.444444444444445,
      "grad_norm": 13.92780590057373,
      "learning_rate": 4.166666666666667e-05,
      "loss": 0.4628,
      "step": 30
    },
    {
      "epoch": 4.888888888888889,
      "eval_accuracy": 0.7455268389662028,
      "eval_loss": 0.5979239344596863,
      "eval_runtime": 2.6102,
      "eval_samples_per_second": 192.706,
      "eval_steps_per_second": 6.13,
      "step": 33
    },
    {
      "epoch": 5.925925925925926,
      "grad_norm": 25.355466842651367,
      "learning_rate": 4.938271604938271e-05,
      "loss": 0.4181,
      "step": 40
    },
    {
      "epoch": 5.925925925925926,
      "eval_accuracy": 0.7614314115308151,
      "eval_loss": 0.5735302567481995,
      "eval_runtime": 2.7235,
      "eval_samples_per_second": 184.688,
      "eval_steps_per_second": 5.875,
      "step": 40
    },
    {
      "epoch": 6.962962962962963,
      "eval_accuracy": 0.7495029821073559,
      "eval_loss": 0.5251761078834534,
      "eval_runtime": 2.662,
      "eval_samples_per_second": 188.956,
      "eval_steps_per_second": 6.011,
      "step": 47
    },
    {
      "epoch": 7.407407407407407,
      "grad_norm": 46.14825439453125,
      "learning_rate": 4.783950617283951e-05,
      "loss": 0.4079,
      "step": 50
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.7475149105367793,
      "eval_loss": 0.5362666249275208,
      "eval_runtime": 2.6877,
      "eval_samples_per_second": 187.151,
      "eval_steps_per_second": 5.953,
      "step": 54
    },
    {
      "epoch": 8.88888888888889,
      "grad_norm": 22.70941925048828,
      "learning_rate": 4.62962962962963e-05,
      "loss": 0.4102,
      "step": 60
    },
    {
      "epoch": 8.88888888888889,
      "eval_accuracy": 0.7495029821073559,
      "eval_loss": 0.5288791060447693,
      "eval_runtime": 2.7002,
      "eval_samples_per_second": 186.285,
      "eval_steps_per_second": 5.926,
      "step": 60
    },
    {
      "epoch": 9.925925925925926,
      "eval_accuracy": 0.7534791252485089,
      "eval_loss": 0.522729754447937,
      "eval_runtime": 2.7539,
      "eval_samples_per_second": 182.652,
      "eval_steps_per_second": 5.81,
      "step": 67
    },
    {
      "epoch": 10.37037037037037,
      "grad_norm": 92.61900329589844,
      "learning_rate": 4.4753086419753084e-05,
      "loss": 0.373,
      "step": 70
    },
    {
      "epoch": 10.962962962962964,
      "eval_accuracy": 0.7773359840954275,
      "eval_loss": 0.46767404675483704,
      "eval_runtime": 2.7441,
      "eval_samples_per_second": 183.301,
      "eval_steps_per_second": 5.831,
      "step": 74
    },
    {
      "epoch": 11.851851851851851,
      "grad_norm": 13.839974403381348,
      "learning_rate": 4.3209876543209875e-05,
      "loss": 0.3639,
      "step": 80
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.7813121272365805,
      "eval_loss": 0.4978141188621521,
      "eval_runtime": 2.7627,
      "eval_samples_per_second": 182.069,
      "eval_steps_per_second": 5.791,
      "step": 81
    },
    {
      "epoch": 12.88888888888889,
      "eval_accuracy": 0.7992047713717694,
      "eval_loss": 0.465084969997406,
      "eval_runtime": 2.7393,
      "eval_samples_per_second": 183.625,
      "eval_steps_per_second": 5.841,
      "step": 87
    },
    {
      "epoch": 13.333333333333334,
      "grad_norm": 12.678291320800781,
      "learning_rate": 4.166666666666667e-05,
      "loss": 0.3779,
      "step": 90
    },
    {
      "epoch": 13.925925925925926,
      "eval_accuracy": 0.7912524850894632,
      "eval_loss": 0.4737919569015503,
      "eval_runtime": 2.7321,
      "eval_samples_per_second": 184.107,
      "eval_steps_per_second": 5.856,
      "step": 94
    },
    {
      "epoch": 14.814814814814815,
      "grad_norm": 13.37376594543457,
      "learning_rate": 4.012345679012346e-05,
      "loss": 0.3476,
      "step": 100
    },
    {
      "epoch": 14.962962962962964,
      "eval_accuracy": 0.8071570576540755,
      "eval_loss": 0.46969088912010193,
      "eval_runtime": 2.704,
      "eval_samples_per_second": 186.017,
      "eval_steps_per_second": 5.917,
      "step": 101
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.7952286282306164,
      "eval_loss": 0.47190144658088684,
      "eval_runtime": 2.651,
      "eval_samples_per_second": 189.741,
      "eval_steps_per_second": 6.036,
      "step": 108
    },
    {
      "epoch": 16.296296296296298,
      "grad_norm": 36.415714263916016,
      "learning_rate": 3.8580246913580246e-05,
      "loss": 0.3467,
      "step": 110
    },
    {
      "epoch": 16.88888888888889,
      "eval_accuracy": 0.7892644135188867,
      "eval_loss": 0.4551805257797241,
      "eval_runtime": 2.7016,
      "eval_samples_per_second": 186.184,
      "eval_steps_per_second": 5.922,
      "step": 114
    },
    {
      "epoch": 17.77777777777778,
      "grad_norm": 21.425716400146484,
      "learning_rate": 3.7037037037037037e-05,
      "loss": 0.3496,
      "step": 120
    },
    {
      "epoch": 17.925925925925927,
      "eval_accuracy": 0.7713717693836978,
      "eval_loss": 0.5186495780944824,
      "eval_runtime": 2.9189,
      "eval_samples_per_second": 172.324,
      "eval_steps_per_second": 5.481,
      "step": 121
    },
    {
      "epoch": 18.962962962962962,
      "eval_accuracy": 0.7952286282306164,
      "eval_loss": 0.4575484097003937,
      "eval_runtime": 2.8214,
      "eval_samples_per_second": 178.277,
      "eval_steps_per_second": 5.671,
      "step": 128
    },
    {
      "epoch": 19.25925925925926,
      "grad_norm": 25.632354736328125,
      "learning_rate": 3.5493827160493834e-05,
      "loss": 0.3657,
      "step": 130
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 0.7793240556660039,
      "eval_loss": 0.476377010345459,
      "eval_runtime": 2.6973,
      "eval_samples_per_second": 186.481,
      "eval_steps_per_second": 5.932,
      "step": 135
    },
    {
      "epoch": 20.74074074074074,
      "grad_norm": 37.57546615600586,
      "learning_rate": 3.395061728395062e-05,
      "loss": 0.3888,
      "step": 140
    },
    {
      "epoch": 20.88888888888889,
      "eval_accuracy": 0.7713717693836978,
      "eval_loss": 0.5008699297904968,
      "eval_runtime": 2.7366,
      "eval_samples_per_second": 183.807,
      "eval_steps_per_second": 5.847,
      "step": 141
    },
    {
      "epoch": 21.925925925925927,
      "eval_accuracy": 0.7813121272365805,
      "eval_loss": 0.46731534600257874,
      "eval_runtime": 2.8548,
      "eval_samples_per_second": 176.192,
      "eval_steps_per_second": 5.605,
      "step": 148
    },
    {
      "epoch": 22.22222222222222,
      "grad_norm": 23.367645263671875,
      "learning_rate": 3.240740740740741e-05,
      "loss": 0.3236,
      "step": 150
    },
    {
      "epoch": 22.962962962962962,
      "eval_accuracy": 0.7753479125248509,
      "eval_loss": 0.493118554353714,
      "eval_runtime": 2.8057,
      "eval_samples_per_second": 179.277,
      "eval_steps_per_second": 5.703,
      "step": 155
    },
    {
      "epoch": 23.703703703703702,
      "grad_norm": 41.31706237792969,
      "learning_rate": 3.08641975308642e-05,
      "loss": 0.3179,
      "step": 160
    },
    {
      "epoch": 24.0,
      "eval_accuracy": 0.7654075546719682,
      "eval_loss": 0.4837174415588379,
      "eval_runtime": 2.8061,
      "eval_samples_per_second": 179.252,
      "eval_steps_per_second": 5.702,
      "step": 162
    },
    {
      "epoch": 24.88888888888889,
      "eval_accuracy": 0.7693836978131213,
      "eval_loss": 0.4651556611061096,
      "eval_runtime": 2.7801,
      "eval_samples_per_second": 180.927,
      "eval_steps_per_second": 5.755,
      "step": 168
    },
    {
      "epoch": 25.185185185185187,
      "grad_norm": 48.94294357299805,
      "learning_rate": 2.9320987654320992e-05,
      "loss": 0.327,
      "step": 170
    },
    {
      "epoch": 25.925925925925927,
      "eval_accuracy": 0.7495029821073559,
      "eval_loss": 0.5107513070106506,
      "eval_runtime": 2.746,
      "eval_samples_per_second": 183.173,
      "eval_steps_per_second": 5.827,
      "step": 175
    },
    {
      "epoch": 26.666666666666668,
      "grad_norm": 17.72180938720703,
      "learning_rate": 2.777777777777778e-05,
      "loss": 0.3253,
      "step": 180
    },
    {
      "epoch": 26.962962962962962,
      "eval_accuracy": 0.7833001988071571,
      "eval_loss": 0.4423621594905853,
      "eval_runtime": 2.851,
      "eval_samples_per_second": 176.432,
      "eval_steps_per_second": 5.612,
      "step": 182
    },
    {
      "epoch": 28.0,
      "eval_accuracy": 0.7335984095427436,
      "eval_loss": 0.5621975064277649,
      "eval_runtime": 2.8441,
      "eval_samples_per_second": 176.859,
      "eval_steps_per_second": 5.626,
      "step": 189
    },
    {
      "epoch": 28.14814814814815,
      "grad_norm": 26.241844177246094,
      "learning_rate": 2.623456790123457e-05,
      "loss": 0.3382,
      "step": 190
    },
    {
      "epoch": 28.88888888888889,
      "eval_accuracy": 0.7693836978131213,
      "eval_loss": 0.5067819952964783,
      "eval_runtime": 2.8675,
      "eval_samples_per_second": 175.415,
      "eval_steps_per_second": 5.58,
      "step": 195
    },
    {
      "epoch": 29.62962962962963,
      "grad_norm": 12.02319049835205,
      "learning_rate": 2.4691358024691357e-05,
      "loss": 0.331,
      "step": 200
    },
    {
      "epoch": 29.925925925925927,
      "eval_accuracy": 0.7693836978131213,
      "eval_loss": 0.45300325751304626,
      "eval_runtime": 2.7755,
      "eval_samples_per_second": 181.232,
      "eval_steps_per_second": 5.765,
      "step": 202
    },
    {
      "epoch": 30.962962962962962,
      "eval_accuracy": 0.731610337972167,
      "eval_loss": 0.5205386877059937,
      "eval_runtime": 2.8057,
      "eval_samples_per_second": 179.276,
      "eval_steps_per_second": 5.703,
      "step": 209
    },
    {
      "epoch": 31.11111111111111,
      "grad_norm": 14.40517807006836,
      "learning_rate": 2.314814814814815e-05,
      "loss": 0.3302,
      "step": 210
    },
    {
      "epoch": 32.0,
      "eval_accuracy": 0.7852882703777336,
      "eval_loss": 0.4385511875152588,
      "eval_runtime": 2.8045,
      "eval_samples_per_second": 179.357,
      "eval_steps_per_second": 5.705,
      "step": 216
    },
    {
      "epoch": 32.592592592592595,
      "grad_norm": 18.574872970581055,
      "learning_rate": 2.1604938271604937e-05,
      "loss": 0.2972,
      "step": 220
    },
    {
      "epoch": 32.888888888888886,
      "eval_accuracy": 0.7773359840954275,
      "eval_loss": 0.5030562281608582,
      "eval_runtime": 2.6994,
      "eval_samples_per_second": 186.336,
      "eval_steps_per_second": 5.927,
      "step": 222
    },
    {
      "epoch": 33.925925925925924,
      "eval_accuracy": 0.757455268389662,
      "eval_loss": 0.49088525772094727,
      "eval_runtime": 2.709,
      "eval_samples_per_second": 185.676,
      "eval_steps_per_second": 5.906,
      "step": 229
    },
    {
      "epoch": 34.074074074074076,
      "grad_norm": 21.681509017944336,
      "learning_rate": 2.006172839506173e-05,
      "loss": 0.3121,
      "step": 230
    },
    {
      "epoch": 34.96296296296296,
      "eval_accuracy": 0.7793240556660039,
      "eval_loss": 0.47658684849739075,
      "eval_runtime": 2.7374,
      "eval_samples_per_second": 183.751,
      "eval_steps_per_second": 5.845,
      "step": 236
    },
    {
      "epoch": 35.55555555555556,
      "grad_norm": 29.247716903686523,
      "learning_rate": 1.8518518518518518e-05,
      "loss": 0.2956,
      "step": 240
    },
    {
      "epoch": 36.0,
      "eval_accuracy": 0.7415506958250497,
      "eval_loss": 0.5262213945388794,
      "eval_runtime": 2.8701,
      "eval_samples_per_second": 175.256,
      "eval_steps_per_second": 5.575,
      "step": 243
    },
    {
      "epoch": 36.888888888888886,
      "eval_accuracy": 0.731610337972167,
      "eval_loss": 0.5373868942260742,
      "eval_runtime": 2.878,
      "eval_samples_per_second": 174.771,
      "eval_steps_per_second": 5.559,
      "step": 249
    },
    {
      "epoch": 37.03703703703704,
      "grad_norm": 13.55726146697998,
      "learning_rate": 1.697530864197531e-05,
      "loss": 0.2947,
      "step": 250
    },
    {
      "epoch": 37.925925925925924,
      "eval_accuracy": 0.7673956262425448,
      "eval_loss": 0.48880261182785034,
      "eval_runtime": 2.7134,
      "eval_samples_per_second": 185.373,
      "eval_steps_per_second": 5.897,
      "step": 256
    },
    {
      "epoch": 38.51851851851852,
      "grad_norm": 8.393943786621094,
      "learning_rate": 1.54320987654321e-05,
      "loss": 0.2662,
      "step": 260
    },
    {
      "epoch": 38.96296296296296,
      "eval_accuracy": 0.7693836978131213,
      "eval_loss": 0.4880698323249817,
      "eval_runtime": 2.7354,
      "eval_samples_per_second": 183.888,
      "eval_steps_per_second": 5.849,
      "step": 263
    },
    {
      "epoch": 40.0,
      "grad_norm": 24.057205200195312,
      "learning_rate": 1.388888888888889e-05,
      "loss": 0.2826,
      "step": 270
    },
    {
      "epoch": 40.0,
      "eval_accuracy": 0.7892644135188867,
      "eval_loss": 0.46687519550323486,
      "eval_runtime": 2.7865,
      "eval_samples_per_second": 180.514,
      "eval_steps_per_second": 5.742,
      "step": 270
    },
    {
      "epoch": 40.888888888888886,
      "eval_accuracy": 0.7972166998011928,
      "eval_loss": 0.45914533734321594,
      "eval_runtime": 2.8041,
      "eval_samples_per_second": 179.381,
      "eval_steps_per_second": 5.706,
      "step": 276
    },
    {
      "epoch": 41.48148148148148,
      "grad_norm": 20.68549346923828,
      "learning_rate": 1.2345679012345678e-05,
      "loss": 0.2768,
      "step": 280
    },
    {
      "epoch": 41.925925925925924,
      "eval_accuracy": 0.757455268389662,
      "eval_loss": 0.5089908838272095,
      "eval_runtime": 2.9736,
      "eval_samples_per_second": 169.157,
      "eval_steps_per_second": 5.381,
      "step": 283
    },
    {
      "epoch": 42.96296296296296,
      "grad_norm": 9.9234619140625,
      "learning_rate": 1.0802469135802469e-05,
      "loss": 0.2836,
      "step": 290
    },
    {
      "epoch": 42.96296296296296,
      "eval_accuracy": 0.7495029821073559,
      "eval_loss": 0.5249876379966736,
      "eval_runtime": 2.8985,
      "eval_samples_per_second": 173.54,
      "eval_steps_per_second": 5.52,
      "step": 290
    },
    {
      "epoch": 44.0,
      "eval_accuracy": 0.7654075546719682,
      "eval_loss": 0.4747855067253113,
      "eval_runtime": 2.9057,
      "eval_samples_per_second": 173.108,
      "eval_steps_per_second": 5.506,
      "step": 297
    },
    {
      "epoch": 44.44444444444444,
      "grad_norm": 19.76637840270996,
      "learning_rate": 9.259259259259259e-06,
      "loss": 0.2724,
      "step": 300
    },
    {
      "epoch": 44.888888888888886,
      "eval_accuracy": 0.7833001988071571,
      "eval_loss": 0.44288724660873413,
      "eval_runtime": 2.7865,
      "eval_samples_per_second": 180.51,
      "eval_steps_per_second": 5.742,
      "step": 303
    },
    {
      "epoch": 45.925925925925924,
      "grad_norm": 12.390064239501953,
      "learning_rate": 7.71604938271605e-06,
      "loss": 0.2498,
      "step": 310
    },
    {
      "epoch": 45.925925925925924,
      "eval_accuracy": 0.7892644135188867,
      "eval_loss": 0.4459961950778961,
      "eval_runtime": 2.8017,
      "eval_samples_per_second": 179.537,
      "eval_steps_per_second": 5.711,
      "step": 310
    },
    {
      "epoch": 46.96296296296296,
      "eval_accuracy": 0.7793240556660039,
      "eval_loss": 0.4721997082233429,
      "eval_runtime": 2.8302,
      "eval_samples_per_second": 177.729,
      "eval_steps_per_second": 5.653,
      "step": 317
    },
    {
      "epoch": 47.407407407407405,
      "grad_norm": 11.559773445129395,
      "learning_rate": 6.172839506172839e-06,
      "loss": 0.2893,
      "step": 320
    },
    {
      "epoch": 48.0,
      "eval_accuracy": 0.7713717693836978,
      "eval_loss": 0.47993555665016174,
      "eval_runtime": 2.9229,
      "eval_samples_per_second": 172.087,
      "eval_steps_per_second": 5.474,
      "step": 324
    },
    {
      "epoch": 48.888888888888886,
      "grad_norm": 11.171250343322754,
      "learning_rate": 4.6296296296296296e-06,
      "loss": 0.2618,
      "step": 330
    },
    {
      "epoch": 48.888888888888886,
      "eval_accuracy": 0.7713717693836978,
      "eval_loss": 0.4849596321582794,
      "eval_runtime": 2.8546,
      "eval_samples_per_second": 176.209,
      "eval_steps_per_second": 5.605,
      "step": 330
    },
    {
      "epoch": 49.925925925925924,
      "eval_accuracy": 0.7495029821073559,
      "eval_loss": 0.5151545405387878,
      "eval_runtime": 2.8154,
      "eval_samples_per_second": 178.658,
      "eval_steps_per_second": 5.683,
      "step": 337
    },
    {
      "epoch": 50.370370370370374,
      "grad_norm": 16.287992477416992,
      "learning_rate": 3.0864197530864196e-06,
      "loss": 0.2664,
      "step": 340
    },
    {
      "epoch": 50.96296296296296,
      "eval_accuracy": 0.7395626242544732,
      "eval_loss": 0.5347036123275757,
      "eval_runtime": 2.8326,
      "eval_samples_per_second": 177.574,
      "eval_steps_per_second": 5.648,
      "step": 344
    },
    {
      "epoch": 51.851851851851855,
      "grad_norm": 15.289400100708008,
      "learning_rate": 1.5432098765432098e-06,
      "loss": 0.27,
      "step": 350
    },
    {
      "epoch": 52.0,
      "eval_accuracy": 0.7415506958250497,
      "eval_loss": 0.5342876315116882,
      "eval_runtime": 2.8365,
      "eval_samples_per_second": 177.329,
      "eval_steps_per_second": 5.641,
      "step": 351
    },
    {
      "epoch": 52.888888888888886,
      "eval_accuracy": 0.7415506958250497,
      "eval_loss": 0.5330411195755005,
      "eval_runtime": 2.823,
      "eval_samples_per_second": 178.181,
      "eval_steps_per_second": 5.668,
      "step": 357
    },
    {
      "epoch": 53.333333333333336,
      "grad_norm": 15.15584945678711,
      "learning_rate": 0.0,
      "loss": 0.2539,
      "step": 360
    },
    {
      "epoch": 53.333333333333336,
      "eval_accuracy": 0.7395626242544732,
      "eval_loss": 0.5319550037384033,
      "eval_runtime": 2.8191,
      "eval_samples_per_second": 178.427,
      "eval_steps_per_second": 5.676,
      "step": 360
    },
    {
      "epoch": 53.333333333333336,
      "step": 360,
      "total_flos": 1.140094502803243e+18,
      "train_loss": 0.34168325927522447,
      "train_runtime": 709.0525,
      "train_samples_per_second": 72.773,
      "train_steps_per_second": 0.508
    }
  ],
  "logging_steps": 10,
  "max_steps": 360,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 60,
  "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": 1.140094502803243e+18,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}