{ "best_metric": 0.0906025692820549, "best_model_checkpoint": "./vit-base-16-thesis-demo-ISIC-multi-class/checkpoint-200", "epoch": 4.0, "eval_steps": 50, "global_step": 204, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.2, "learning_rate": 0.00019019607843137254, "loss": 0.9473, "step": 10 }, { "epoch": 0.39, "learning_rate": 0.0001803921568627451, "loss": 0.6725, "step": 20 }, { "epoch": 0.59, "learning_rate": 0.00017058823529411766, "loss": 0.6139, "step": 30 }, { "epoch": 0.78, "learning_rate": 0.00016078431372549022, "loss": 0.5606, "step": 40 }, { "epoch": 0.98, "learning_rate": 0.00015196078431372549, "loss": 0.575, "step": 50 }, { "epoch": 0.98, "eval_accuracy": 0.8490566037735849, "eval_f1": 0.8490566037735849, "eval_loss": 0.4132387042045593, "eval_precision": 0.8490566037735849, "eval_recall": 0.8490566037735849, "eval_runtime": 20.7397, "eval_samples_per_second": 7.666, "eval_steps_per_second": 0.964, "step": 50 }, { "epoch": 1.18, "learning_rate": 0.00014215686274509804, "loss": 0.5265, "step": 60 }, { "epoch": 1.37, "learning_rate": 0.0001323529411764706, "loss": 0.4456, "step": 70 }, { "epoch": 1.57, "learning_rate": 0.00012254901960784316, "loss": 0.3044, "step": 80 }, { "epoch": 1.76, "learning_rate": 0.0001127450980392157, "loss": 0.3829, "step": 90 }, { "epoch": 1.96, "learning_rate": 0.00010294117647058823, "loss": 0.2771, "step": 100 }, { "epoch": 1.96, "eval_accuracy": 0.9182389937106918, "eval_f1": 0.9182389937106918, "eval_loss": 0.23294508457183838, "eval_precision": 0.9182389937106918, "eval_recall": 0.9182389937106918, "eval_runtime": 18.6716, "eval_samples_per_second": 8.516, "eval_steps_per_second": 1.071, "step": 100 }, { "epoch": 2.16, "learning_rate": 9.313725490196079e-05, "loss": 0.2781, "step": 110 }, { "epoch": 2.35, "learning_rate": 8.333333333333334e-05, "loss": 0.2399, "step": 120 }, { "epoch": 2.55, "learning_rate": 7.352941176470589e-05, "loss": 0.1721, "step": 130 }, { "epoch": 2.75, "learning_rate": 6.372549019607843e-05, "loss": 0.1759, "step": 140 }, { "epoch": 2.94, "learning_rate": 5.392156862745098e-05, "loss": 0.1703, "step": 150 }, { "epoch": 2.94, "eval_accuracy": 0.949685534591195, "eval_f1": 0.949685534591195, "eval_loss": 0.18214598298072815, "eval_precision": 0.949685534591195, "eval_recall": 0.949685534591195, "eval_runtime": 19.3259, "eval_samples_per_second": 8.227, "eval_steps_per_second": 1.035, "step": 150 }, { "epoch": 3.14, "learning_rate": 4.411764705882353e-05, "loss": 0.0989, "step": 160 }, { "epoch": 3.33, "learning_rate": 3.431372549019608e-05, "loss": 0.0797, "step": 170 }, { "epoch": 3.53, "learning_rate": 2.4509803921568626e-05, "loss": 0.1056, "step": 180 }, { "epoch": 3.73, "learning_rate": 1.4705882352941177e-05, "loss": 0.0492, "step": 190 }, { "epoch": 3.92, "learning_rate": 4.901960784313726e-06, "loss": 0.1186, "step": 200 }, { "epoch": 3.92, "eval_accuracy": 0.9748427672955975, "eval_f1": 0.9748427672955975, "eval_loss": 0.0906025692820549, "eval_precision": 0.9748427672955975, "eval_recall": 0.9748427672955975, "eval_runtime": 18.0737, "eval_samples_per_second": 8.797, "eval_steps_per_second": 1.107, "step": 200 }, { "epoch": 4.0, "step": 204, "total_flos": 2.5014638443873075e+17, "train_loss": 0.33582471369528305, "train_runtime": 621.9278, "train_samples_per_second": 5.19, "train_steps_per_second": 0.328 } ], "logging_steps": 10, "max_steps": 204, "num_train_epochs": 4, "save_steps": 50, "total_flos": 2.5014638443873075e+17, "trial_name": null, "trial_params": null }