{
  "best_metric": 0.0827358290553093,
  "best_model_checkpoint": "./vit-large-modified-augmented-ph2-patch-16/checkpoint-650",
  "epoch": 4.0,
  "global_step": 680,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.06,
      "learning_rate": 0.00019705882352941177,
      "loss": 0.6824,
      "step": 10
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.00019441176470588235,
      "loss": 0.4985,
      "step": 20
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.00019147058823529414,
      "loss": 0.4406,
      "step": 30
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.00018852941176470588,
      "loss": 0.4613,
      "step": 40
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.00018558823529411767,
      "loss": 0.3402,
      "step": 50
    },
    {
      "epoch": 0.29,
      "eval_accuracy": 0.7945205479452054,
      "eval_f1": 0.7945205479452053,
      "eval_loss": 0.6269038915634155,
      "eval_precision": 0.7945205479452054,
      "eval_recall": 0.7945205479452054,
      "eval_runtime": 11.3009,
      "eval_samples_per_second": 51.677,
      "eval_steps_per_second": 6.46,
      "step": 50
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.0001826470588235294,
      "loss": 0.4146,
      "step": 60
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.0001797058823529412,
      "loss": 0.3038,
      "step": 70
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00017676470588235294,
      "loss": 0.2719,
      "step": 80
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.00017382352941176473,
      "loss": 0.3276,
      "step": 90
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00017088235294117647,
      "loss": 0.1387,
      "step": 100
    },
    {
      "epoch": 0.59,
      "eval_accuracy": 0.8921232876712328,
      "eval_f1": 0.8921232876712328,
      "eval_loss": 0.2956628203392029,
      "eval_precision": 0.8921232876712328,
      "eval_recall": 0.8921232876712328,
      "eval_runtime": 11.8234,
      "eval_samples_per_second": 49.393,
      "eval_steps_per_second": 6.174,
      "step": 100
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.00016794117647058823,
      "loss": 0.2664,
      "step": 110
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.000165,
      "loss": 0.2601,
      "step": 120
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.00016205882352941176,
      "loss": 0.2548,
      "step": 130
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.00015911764705882353,
      "loss": 0.3336,
      "step": 140
    },
    {
      "epoch": 0.88,
      "learning_rate": 0.0001561764705882353,
      "loss": 0.2921,
      "step": 150
    },
    {
      "epoch": 0.88,
      "eval_accuracy": 0.8835616438356164,
      "eval_f1": 0.8835616438356163,
      "eval_loss": 0.31567689776420593,
      "eval_precision": 0.8835616438356164,
      "eval_recall": 0.8835616438356164,
      "eval_runtime": 11.8613,
      "eval_samples_per_second": 49.236,
      "eval_steps_per_second": 6.154,
      "step": 150
    },
    {
      "epoch": 0.94,
      "learning_rate": 0.00015323529411764709,
      "loss": 0.2107,
      "step": 160
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.00015029411764705882,
      "loss": 0.1724,
      "step": 170
    },
    {
      "epoch": 1.06,
      "learning_rate": 0.00014735294117647062,
      "loss": 0.1688,
      "step": 180
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.00014441176470588235,
      "loss": 0.2278,
      "step": 190
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.00014147058823529415,
      "loss": 0.1268,
      "step": 200
    },
    {
      "epoch": 1.18,
      "eval_accuracy": 0.8527397260273972,
      "eval_f1": 0.8527397260273972,
      "eval_loss": 0.455745130777359,
      "eval_precision": 0.8527397260273972,
      "eval_recall": 0.8527397260273972,
      "eval_runtime": 11.8688,
      "eval_samples_per_second": 49.205,
      "eval_steps_per_second": 6.151,
      "step": 200
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.00013852941176470588,
      "loss": 0.2317,
      "step": 210
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.00013558823529411765,
      "loss": 0.1221,
      "step": 220
    },
    {
      "epoch": 1.35,
      "learning_rate": 0.00013264705882352941,
      "loss": 0.1903,
      "step": 230
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.00012970588235294118,
      "loss": 0.1622,
      "step": 240
    },
    {
      "epoch": 1.47,
      "learning_rate": 0.00012676470588235294,
      "loss": 0.2071,
      "step": 250
    },
    {
      "epoch": 1.47,
      "eval_accuracy": 0.8818493150684932,
      "eval_f1": 0.8818493150684932,
      "eval_loss": 0.26901647448539734,
      "eval_precision": 0.8818493150684932,
      "eval_recall": 0.8818493150684932,
      "eval_runtime": 12.0775,
      "eval_samples_per_second": 48.354,
      "eval_steps_per_second": 6.044,
      "step": 250
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.0001238235294117647,
      "loss": 0.265,
      "step": 260
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.00012088235294117647,
      "loss": 0.215,
      "step": 270
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.00011794117647058824,
      "loss": 0.1359,
      "step": 280
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.00011499999999999999,
      "loss": 0.1603,
      "step": 290
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.00011205882352941177,
      "loss": 0.1238,
      "step": 300
    },
    {
      "epoch": 1.76,
      "eval_accuracy": 0.9178082191780822,
      "eval_f1": 0.9178082191780822,
      "eval_loss": 0.29986169934272766,
      "eval_precision": 0.9178082191780822,
      "eval_recall": 0.9178082191780822,
      "eval_runtime": 12.0554,
      "eval_samples_per_second": 48.443,
      "eval_steps_per_second": 6.055,
      "step": 300
    },
    {
      "epoch": 1.82,
      "learning_rate": 0.00010911764705882352,
      "loss": 0.097,
      "step": 310
    },
    {
      "epoch": 1.88,
      "learning_rate": 0.0001061764705882353,
      "loss": 0.244,
      "step": 320
    },
    {
      "epoch": 1.94,
      "learning_rate": 0.00010323529411764708,
      "loss": 0.1553,
      "step": 330
    },
    {
      "epoch": 2.0,
      "learning_rate": 0.00010029411764705883,
      "loss": 0.0799,
      "step": 340
    },
    {
      "epoch": 2.06,
      "learning_rate": 9.73529411764706e-05,
      "loss": 0.1327,
      "step": 350
    },
    {
      "epoch": 2.06,
      "eval_accuracy": 0.7876712328767124,
      "eval_f1": 0.7876712328767124,
      "eval_loss": 0.6025873422622681,
      "eval_precision": 0.7876712328767124,
      "eval_recall": 0.7876712328767124,
      "eval_runtime": 12.0601,
      "eval_samples_per_second": 48.424,
      "eval_steps_per_second": 6.053,
      "step": 350
    },
    {
      "epoch": 2.12,
      "learning_rate": 9.441176470588236e-05,
      "loss": 0.1859,
      "step": 360
    },
    {
      "epoch": 2.18,
      "learning_rate": 9.147058823529412e-05,
      "loss": 0.1577,
      "step": 370
    },
    {
      "epoch": 2.24,
      "learning_rate": 8.852941176470589e-05,
      "loss": 0.1616,
      "step": 380
    },
    {
      "epoch": 2.29,
      "learning_rate": 8.558823529411765e-05,
      "loss": 0.1998,
      "step": 390
    },
    {
      "epoch": 2.35,
      "learning_rate": 8.26470588235294e-05,
      "loss": 0.1453,
      "step": 400
    },
    {
      "epoch": 2.35,
      "eval_accuracy": 0.898972602739726,
      "eval_f1": 0.8989726027397261,
      "eval_loss": 0.2887475788593292,
      "eval_precision": 0.898972602739726,
      "eval_recall": 0.898972602739726,
      "eval_runtime": 11.7988,
      "eval_samples_per_second": 49.497,
      "eval_steps_per_second": 6.187,
      "step": 400
    },
    {
      "epoch": 2.41,
      "learning_rate": 7.970588235294118e-05,
      "loss": 0.1919,
      "step": 410
    },
    {
      "epoch": 2.47,
      "learning_rate": 7.676470588235295e-05,
      "loss": 0.0973,
      "step": 420
    },
    {
      "epoch": 2.53,
      "learning_rate": 7.382352941176471e-05,
      "loss": 0.0649,
      "step": 430
    },
    {
      "epoch": 2.59,
      "learning_rate": 7.088235294117648e-05,
      "loss": 0.1033,
      "step": 440
    },
    {
      "epoch": 2.65,
      "learning_rate": 6.794117647058824e-05,
      "loss": 0.0686,
      "step": 450
    },
    {
      "epoch": 2.65,
      "eval_accuracy": 0.9503424657534246,
      "eval_f1": 0.9503424657534246,
      "eval_loss": 0.2048829048871994,
      "eval_precision": 0.9503424657534246,
      "eval_recall": 0.9503424657534246,
      "eval_runtime": 11.8436,
      "eval_samples_per_second": 49.309,
      "eval_steps_per_second": 6.164,
      "step": 450
    },
    {
      "epoch": 2.71,
      "learning_rate": 6.500000000000001e-05,
      "loss": 0.0378,
      "step": 460
    },
    {
      "epoch": 2.76,
      "learning_rate": 6.205882352941177e-05,
      "loss": 0.095,
      "step": 470
    },
    {
      "epoch": 2.82,
      "learning_rate": 5.911764705882353e-05,
      "loss": 0.0501,
      "step": 480
    },
    {
      "epoch": 2.88,
      "learning_rate": 5.6176470588235296e-05,
      "loss": 0.0444,
      "step": 490
    },
    {
      "epoch": 2.94,
      "learning_rate": 5.323529411764706e-05,
      "loss": 0.0414,
      "step": 500
    },
    {
      "epoch": 2.94,
      "eval_accuracy": 0.9195205479452054,
      "eval_f1": 0.9195205479452054,
      "eval_loss": 0.30401644110679626,
      "eval_precision": 0.9195205479452054,
      "eval_recall": 0.9195205479452054,
      "eval_runtime": 11.748,
      "eval_samples_per_second": 49.711,
      "eval_steps_per_second": 6.214,
      "step": 500
    },
    {
      "epoch": 3.0,
      "learning_rate": 5.0294117647058826e-05,
      "loss": 0.0455,
      "step": 510
    },
    {
      "epoch": 3.06,
      "learning_rate": 4.735294117647059e-05,
      "loss": 0.0631,
      "step": 520
    },
    {
      "epoch": 3.12,
      "learning_rate": 4.4411764705882356e-05,
      "loss": 0.1279,
      "step": 530
    },
    {
      "epoch": 3.18,
      "learning_rate": 4.147058823529412e-05,
      "loss": 0.0111,
      "step": 540
    },
    {
      "epoch": 3.24,
      "learning_rate": 3.8529411764705886e-05,
      "loss": 0.0851,
      "step": 550
    },
    {
      "epoch": 3.24,
      "eval_accuracy": 0.9297945205479452,
      "eval_f1": 0.9297945205479452,
      "eval_loss": 0.22443008422851562,
      "eval_precision": 0.9297945205479452,
      "eval_recall": 0.9297945205479452,
      "eval_runtime": 11.6193,
      "eval_samples_per_second": 50.261,
      "eval_steps_per_second": 6.283,
      "step": 550
    },
    {
      "epoch": 3.29,
      "learning_rate": 3.558823529411765e-05,
      "loss": 0.0775,
      "step": 560
    },
    {
      "epoch": 3.35,
      "learning_rate": 3.2647058823529416e-05,
      "loss": 0.0075,
      "step": 570
    },
    {
      "epoch": 3.41,
      "learning_rate": 2.9705882352941177e-05,
      "loss": 0.0334,
      "step": 580
    },
    {
      "epoch": 3.47,
      "learning_rate": 2.6764705882352942e-05,
      "loss": 0.0253,
      "step": 590
    },
    {
      "epoch": 3.53,
      "learning_rate": 2.3823529411764707e-05,
      "loss": 0.0054,
      "step": 600
    },
    {
      "epoch": 3.53,
      "eval_accuracy": 0.9554794520547946,
      "eval_f1": 0.9554794520547946,
      "eval_loss": 0.13558290898799896,
      "eval_precision": 0.9554794520547946,
      "eval_recall": 0.9554794520547946,
      "eval_runtime": 11.7366,
      "eval_samples_per_second": 49.759,
      "eval_steps_per_second": 6.22,
      "step": 600
    },
    {
      "epoch": 3.59,
      "learning_rate": 2.0882352941176472e-05,
      "loss": 0.0032,
      "step": 610
    },
    {
      "epoch": 3.65,
      "learning_rate": 1.7941176470588237e-05,
      "loss": 0.0122,
      "step": 620
    },
    {
      "epoch": 3.71,
      "learning_rate": 1.5e-05,
      "loss": 0.0746,
      "step": 630
    },
    {
      "epoch": 3.76,
      "learning_rate": 1.2058823529411765e-05,
      "loss": 0.0416,
      "step": 640
    },
    {
      "epoch": 3.82,
      "learning_rate": 9.117647058823529e-06,
      "loss": 0.0029,
      "step": 650
    },
    {
      "epoch": 3.82,
      "eval_accuracy": 0.9708904109589042,
      "eval_f1": 0.9708904109589042,
      "eval_loss": 0.0827358290553093,
      "eval_precision": 0.9708904109589042,
      "eval_recall": 0.9708904109589042,
      "eval_runtime": 11.7473,
      "eval_samples_per_second": 49.713,
      "eval_steps_per_second": 6.214,
      "step": 650
    },
    {
      "epoch": 3.88,
      "learning_rate": 6.1764705882352944e-06,
      "loss": 0.012,
      "step": 660
    },
    {
      "epoch": 3.94,
      "learning_rate": 3.2352941176470594e-06,
      "loss": 0.0052,
      "step": 670
    },
    {
      "epoch": 4.0,
      "learning_rate": 2.9411764705882356e-07,
      "loss": 0.0441,
      "step": 680
    },
    {
      "epoch": 4.0,
      "step": 680,
      "total_flos": 2.973829456609837e+18,
      "train_loss": 0.16232660566182697,
      "train_runtime": 690.9367,
      "train_samples_per_second": 15.712,
      "train_steps_per_second": 0.984
    }
  ],
  "max_steps": 680,
  "num_train_epochs": 4,
  "total_flos": 2.973829456609837e+18,
  "trial_name": null,
  "trial_params": null
}