{
  "best_metric": 0.8647342995169082,
  "best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-eurosat/checkpoint-234",
  "epoch": 20.0,
  "eval_steps": 500,
  "global_step": 260,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.7692307692307693,
      "grad_norm": 6.725746154785156,
      "learning_rate": 1.923076923076923e-05,
      "loss": 1.7993,
      "step": 10
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5942028985507246,
      "eval_loss": 1.321716070175171,
      "eval_runtime": 3.305,
      "eval_samples_per_second": 62.633,
      "eval_steps_per_second": 2.118,
      "step": 13
    },
    {
      "epoch": 1.5384615384615383,
      "grad_norm": 6.603445053100586,
      "learning_rate": 3.846153846153846e-05,
      "loss": 1.23,
      "step": 20
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6570048309178744,
      "eval_loss": 0.9921989440917969,
      "eval_runtime": 2.4985,
      "eval_samples_per_second": 82.85,
      "eval_steps_per_second": 2.802,
      "step": 26
    },
    {
      "epoch": 2.3076923076923075,
      "grad_norm": 6.3381571769714355,
      "learning_rate": 4.9145299145299147e-05,
      "loss": 1.0516,
      "step": 30
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6231884057971014,
      "eval_loss": 1.0678304433822632,
      "eval_runtime": 2.4205,
      "eval_samples_per_second": 85.52,
      "eval_steps_per_second": 2.892,
      "step": 39
    },
    {
      "epoch": 3.076923076923077,
      "grad_norm": 7.780747413635254,
      "learning_rate": 4.700854700854701e-05,
      "loss": 0.8173,
      "step": 40
    },
    {
      "epoch": 3.8461538461538463,
      "grad_norm": 7.331319808959961,
      "learning_rate": 4.4871794871794874e-05,
      "loss": 0.7172,
      "step": 50
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6618357487922706,
      "eval_loss": 0.8940783739089966,
      "eval_runtime": 2.4149,
      "eval_samples_per_second": 85.718,
      "eval_steps_per_second": 2.899,
      "step": 52
    },
    {
      "epoch": 4.615384615384615,
      "grad_norm": 8.189846992492676,
      "learning_rate": 4.2735042735042735e-05,
      "loss": 0.7044,
      "step": 60
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.782608695652174,
      "eval_loss": 0.6553784608840942,
      "eval_runtime": 2.4376,
      "eval_samples_per_second": 84.918,
      "eval_steps_per_second": 2.872,
      "step": 65
    },
    {
      "epoch": 5.384615384615385,
      "grad_norm": 5.5380377769470215,
      "learning_rate": 4.05982905982906e-05,
      "loss": 0.5443,
      "step": 70
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.7536231884057971,
      "eval_loss": 0.6407715678215027,
      "eval_runtime": 2.4636,
      "eval_samples_per_second": 84.022,
      "eval_steps_per_second": 2.841,
      "step": 78
    },
    {
      "epoch": 6.153846153846154,
      "grad_norm": 4.97896671295166,
      "learning_rate": 3.846153846153846e-05,
      "loss": 0.5464,
      "step": 80
    },
    {
      "epoch": 6.923076923076923,
      "grad_norm": 7.994241714477539,
      "learning_rate": 3.6324786324786323e-05,
      "loss": 0.516,
      "step": 90
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8260869565217391,
      "eval_loss": 0.5527260899543762,
      "eval_runtime": 2.4157,
      "eval_samples_per_second": 85.688,
      "eval_steps_per_second": 2.898,
      "step": 91
    },
    {
      "epoch": 7.6923076923076925,
      "grad_norm": 7.403011322021484,
      "learning_rate": 3.418803418803419e-05,
      "loss": 0.4639,
      "step": 100
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8357487922705314,
      "eval_loss": 0.5045546889305115,
      "eval_runtime": 2.4127,
      "eval_samples_per_second": 85.798,
      "eval_steps_per_second": 2.901,
      "step": 104
    },
    {
      "epoch": 8.461538461538462,
      "grad_norm": 6.654711723327637,
      "learning_rate": 3.205128205128206e-05,
      "loss": 0.5129,
      "step": 110
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.8019323671497585,
      "eval_loss": 0.5691552758216858,
      "eval_runtime": 2.4269,
      "eval_samples_per_second": 85.294,
      "eval_steps_per_second": 2.884,
      "step": 117
    },
    {
      "epoch": 9.23076923076923,
      "grad_norm": 5.1390228271484375,
      "learning_rate": 2.9914529914529915e-05,
      "loss": 0.4434,
      "step": 120
    },
    {
      "epoch": 10.0,
      "grad_norm": 7.468532562255859,
      "learning_rate": 2.777777777777778e-05,
      "loss": 0.4723,
      "step": 130
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.7874396135265701,
      "eval_loss": 0.5761130452156067,
      "eval_runtime": 2.5844,
      "eval_samples_per_second": 80.097,
      "eval_steps_per_second": 2.709,
      "step": 130
    },
    {
      "epoch": 10.76923076923077,
      "grad_norm": 5.666627407073975,
      "learning_rate": 2.564102564102564e-05,
      "loss": 0.3949,
      "step": 140
    },
    {
      "epoch": 11.0,
      "eval_accuracy": 0.7729468599033816,
      "eval_loss": 0.5835833549499512,
      "eval_runtime": 2.4621,
      "eval_samples_per_second": 84.075,
      "eval_steps_per_second": 2.843,
      "step": 143
    },
    {
      "epoch": 11.538461538461538,
      "grad_norm": 3.5363669395446777,
      "learning_rate": 2.3504273504273504e-05,
      "loss": 0.3694,
      "step": 150
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.8115942028985508,
      "eval_loss": 0.5314078330993652,
      "eval_runtime": 2.5082,
      "eval_samples_per_second": 82.528,
      "eval_steps_per_second": 2.791,
      "step": 156
    },
    {
      "epoch": 12.307692307692308,
      "grad_norm": 13.049053192138672,
      "learning_rate": 2.1367521367521368e-05,
      "loss": 0.4252,
      "step": 160
    },
    {
      "epoch": 13.0,
      "eval_accuracy": 0.8502415458937198,
      "eval_loss": 0.45385822653770447,
      "eval_runtime": 2.4622,
      "eval_samples_per_second": 84.072,
      "eval_steps_per_second": 2.843,
      "step": 169
    },
    {
      "epoch": 13.076923076923077,
      "grad_norm": 5.0738983154296875,
      "learning_rate": 1.923076923076923e-05,
      "loss": 0.3531,
      "step": 170
    },
    {
      "epoch": 13.846153846153847,
      "grad_norm": 4.928651332855225,
      "learning_rate": 1.7094017094017095e-05,
      "loss": 0.3532,
      "step": 180
    },
    {
      "epoch": 14.0,
      "eval_accuracy": 0.8309178743961353,
      "eval_loss": 0.4720558226108551,
      "eval_runtime": 2.455,
      "eval_samples_per_second": 84.317,
      "eval_steps_per_second": 2.851,
      "step": 182
    },
    {
      "epoch": 14.615384615384615,
      "grad_norm": 7.16459321975708,
      "learning_rate": 1.4957264957264958e-05,
      "loss": 0.3556,
      "step": 190
    },
    {
      "epoch": 15.0,
      "eval_accuracy": 0.855072463768116,
      "eval_loss": 0.42704012989997864,
      "eval_runtime": 2.5414,
      "eval_samples_per_second": 81.451,
      "eval_steps_per_second": 2.754,
      "step": 195
    },
    {
      "epoch": 15.384615384615385,
      "grad_norm": 7.167410373687744,
      "learning_rate": 1.282051282051282e-05,
      "loss": 0.3565,
      "step": 200
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.8599033816425121,
      "eval_loss": 0.45548996329307556,
      "eval_runtime": 2.4294,
      "eval_samples_per_second": 85.206,
      "eval_steps_per_second": 2.881,
      "step": 208
    },
    {
      "epoch": 16.153846153846153,
      "grad_norm": 5.148393630981445,
      "learning_rate": 1.0683760683760684e-05,
      "loss": 0.3701,
      "step": 210
    },
    {
      "epoch": 16.923076923076923,
      "grad_norm": 6.511909008026123,
      "learning_rate": 8.547008547008548e-06,
      "loss": 0.3369,
      "step": 220
    },
    {
      "epoch": 17.0,
      "eval_accuracy": 0.8502415458937198,
      "eval_loss": 0.4133504629135132,
      "eval_runtime": 2.4571,
      "eval_samples_per_second": 84.246,
      "eval_steps_per_second": 2.849,
      "step": 221
    },
    {
      "epoch": 17.692307692307693,
      "grad_norm": 4.781186580657959,
      "learning_rate": 6.41025641025641e-06,
      "loss": 0.347,
      "step": 230
    },
    {
      "epoch": 18.0,
      "eval_accuracy": 0.8647342995169082,
      "eval_loss": 0.4232899248600006,
      "eval_runtime": 2.4497,
      "eval_samples_per_second": 84.5,
      "eval_steps_per_second": 2.857,
      "step": 234
    },
    {
      "epoch": 18.46153846153846,
      "grad_norm": 5.714421272277832,
      "learning_rate": 4.273504273504274e-06,
      "loss": 0.3386,
      "step": 240
    },
    {
      "epoch": 19.0,
      "eval_accuracy": 0.8599033816425121,
      "eval_loss": 0.4154907464981079,
      "eval_runtime": 2.5896,
      "eval_samples_per_second": 79.936,
      "eval_steps_per_second": 2.703,
      "step": 247
    },
    {
      "epoch": 19.23076923076923,
      "grad_norm": 6.739828109741211,
      "learning_rate": 2.136752136752137e-06,
      "loss": 0.3366,
      "step": 250
    },
    {
      "epoch": 20.0,
      "grad_norm": 8.137150764465332,
      "learning_rate": 0.0,
      "loss": 0.3312,
      "step": 260
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 0.8647342995169082,
      "eval_loss": 0.4072469174861908,
      "eval_runtime": 2.554,
      "eval_samples_per_second": 81.051,
      "eval_steps_per_second": 2.741,
      "step": 260
    },
    {
      "epoch": 20.0,
      "step": 260,
      "total_flos": 8.20362235004928e+17,
      "train_loss": 0.5571963181862465,
      "train_runtime": 609.4825,
      "train_samples_per_second": 54.144,
      "train_steps_per_second": 0.427
    }
  ],
  "logging_steps": 10,
  "max_steps": 260,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 20,
  "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": 8.20362235004928e+17,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}