{ "best_metric": 0.8819264629725531, "best_model_checkpoint": "swin-base-patch4-window7-224-in22k-cons_Classification_base_V10/checkpoint-105", "epoch": 6.885245901639344, "eval_steps": 500, "global_step": 105, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.6557377049180327, "grad_norm": 3.7154650688171387, "learning_rate": 4.545454545454546e-05, "loss": 1.0279, "step": 10 }, { "epoch": 0.9836065573770492, "step": 15, "train_accuracy": 0.7968204730515703, "train_auc_class_0": 0.8674857214060812, "train_auc_class_1": 0.9047321713989991, "train_auc_class_2": 0.8256152717071332, "train_auc_overall": 0.8659443881707377, "train_loss": 0.5398309230804443, "train_runtime": 704.5916, "train_samples_per_second": 10.981, "train_steps_per_second": 0.087 }, { "epoch": 0.9836065573770492, "eval_accuracy": 0.8234075608493009, "eval_auc_class_0": 0.8869781099574333, "eval_auc_class_1": 0.9183382321313357, "eval_auc_class_2": 0.8605897655254291, "eval_auc_overall": 0.8886353692047327, "eval_loss": 0.4998205900192261, "eval_runtime": 206.129, "eval_samples_per_second": 9.368, "eval_steps_per_second": 0.078, "step": 15 }, { "epoch": 1.3114754098360657, "grad_norm": 1.7328158617019653, "learning_rate": 4.5212765957446815e-05, "loss": 0.5553, "step": 20 }, { "epoch": 1.9672131147540983, "grad_norm": 2.4891183376312256, "learning_rate": 3.9893617021276594e-05, "loss": 0.428, "step": 30 }, { "epoch": 1.9672131147540983, "step": 30, "train_accuracy": 0.8445133772780148, "train_auc_class_0": 0.9255038035954841, "train_auc_class_1": 0.9439145849287703, "train_auc_class_2": 0.9234406921755834, "train_auc_overall": 0.930953026899946, "train_loss": 0.38005331158638, "train_runtime": 696.7728, "train_samples_per_second": 11.104, "train_steps_per_second": 0.088 }, { "epoch": 1.9672131147540983, "eval_accuracy": 0.8493008803728638, "eval_auc_class_0": 0.9309048911962445, "eval_auc_class_1": 0.9438927521686142, "eval_auc_class_2": 0.9451350454807339, "eval_auc_overall": 0.9399775629485309, "eval_loss": 0.373575896024704, "eval_runtime": 206.7801, "eval_samples_per_second": 9.338, "eval_steps_per_second": 0.077, "step": 30 }, { "epoch": 2.6229508196721314, "grad_norm": 1.9731707572937012, "learning_rate": 3.4574468085106386e-05, "loss": 0.3805, "step": 40 }, { "epoch": 2.9508196721311473, "step": 45, "train_accuracy": 0.8635129895308259, "train_auc_class_0": 0.9433487526029413, "train_auc_class_1": 0.9545073030991414, "train_auc_class_2": 0.9527349780371537, "train_auc_overall": 0.9501970112464121, "train_loss": 0.33183473348617554, "train_runtime": 694.8396, "train_samples_per_second": 11.135, "train_steps_per_second": 0.088 }, { "epoch": 2.9508196721311473, "eval_accuracy": 0.8638011393060591, "eval_auc_class_0": 0.9430968293750248, "eval_auc_class_1": 0.9533629650871029, "eval_auc_class_2": 0.9590411508982662, "eval_auc_overall": 0.9518336484534647, "eval_loss": 0.3269486129283905, "eval_runtime": 209.5765, "eval_samples_per_second": 9.214, "eval_steps_per_second": 0.076, "step": 45 }, { "epoch": 3.278688524590164, "grad_norm": 2.599257469177246, "learning_rate": 2.925531914893617e-05, "loss": 0.338, "step": 50 }, { "epoch": 3.9344262295081966, "grad_norm": 2.2009198665618896, "learning_rate": 2.393617021276596e-05, "loss": 0.3305, "step": 60 }, { "epoch": 4.0, "step": 61, "train_accuracy": 0.8725604239369265, "train_auc_class_0": 0.948471525573143, "train_auc_class_1": 0.9603955056035058, "train_auc_class_2": 0.9486343423490885, "train_auc_overall": 0.9525004578419124, "train_loss": 0.31600990891456604, "train_runtime": 726.5744, "train_samples_per_second": 10.649, "train_steps_per_second": 0.084 }, { "epoch": 4.0, "eval_accuracy": 0.8762299326773693, "eval_auc_class_0": 0.9471881588892868, "eval_auc_class_1": 0.9563538604917915, "eval_auc_class_2": 0.9630043472947257, "eval_auc_overall": 0.9555154555586013, "eval_loss": 0.3176056146621704, "eval_runtime": 216.0268, "eval_samples_per_second": 8.939, "eval_steps_per_second": 0.074, "step": 61 }, { "epoch": 4.590163934426229, "grad_norm": 2.0908782482147217, "learning_rate": 1.8617021276595745e-05, "loss": 0.3332, "step": 70 }, { "epoch": 4.983606557377049, "step": 76, "train_accuracy": 0.8768256430140882, "train_auc_class_0": 0.9538384356872789, "train_auc_class_1": 0.963852382784682, "train_auc_class_2": 0.9602902514264319, "train_auc_overall": 0.9593270232994642, "train_loss": 0.30048099160194397, "train_runtime": 713.1172, "train_samples_per_second": 10.85, "train_steps_per_second": 0.086 }, { "epoch": 4.983606557377049, "eval_accuracy": 0.8741584671154842, "eval_auc_class_0": 0.946992978477941, "eval_auc_class_1": 0.9557961144168041, "eval_auc_class_2": 0.9635150234823752, "eval_auc_overall": 0.9554347054590401, "eval_loss": 0.31987833976745605, "eval_runtime": 212.2374, "eval_samples_per_second": 9.098, "eval_steps_per_second": 0.075, "step": 76 }, { "epoch": 5.245901639344262, "grad_norm": 2.1055748462677, "learning_rate": 1.3297872340425532e-05, "loss": 0.3097, "step": 80 }, { "epoch": 5.901639344262295, "grad_norm": 1.8331685066223145, "learning_rate": 7.97872340425532e-06, "loss": 0.2986, "step": 90 }, { "epoch": 5.967213114754099, "step": 91, "train_accuracy": 0.8836758433501357, "train_auc_class_0": 0.9567022523385401, "train_auc_class_1": 0.9669396820592466, "train_auc_class_2": 0.9649150652373859, "train_auc_overall": 0.9628523332117241, "train_loss": 0.2871149480342865, "train_runtime": 707.266, "train_samples_per_second": 10.939, "train_steps_per_second": 0.086 }, { "epoch": 5.967213114754099, "eval_accuracy": 0.8767477990678405, "eval_auc_class_0": 0.9486936587500497, "eval_auc_class_1": 0.9580586421965732, "eval_auc_class_2": 0.9654660684044206, "eval_auc_overall": 0.9574061231170146, "eval_loss": 0.3098297417163849, "eval_runtime": 211.4462, "eval_samples_per_second": 9.132, "eval_steps_per_second": 0.076, "step": 91 }, { "epoch": 6.557377049180328, "grad_norm": 1.698627233505249, "learning_rate": 2.6595744680851065e-06, "loss": 0.2891, "step": 100 }, { "epoch": 6.885245901639344, "step": 105, "train_accuracy": 0.8938865193227349, "train_auc_class_0": 0.9620290605729788, "train_auc_class_1": 0.9704803672023041, "train_auc_class_2": 0.967734557500875, "train_auc_overall": 0.9667479950920526, "train_loss": 0.2710348963737488, "train_runtime": 695.2399, "train_samples_per_second": 11.129, "train_steps_per_second": 0.088 }, { "epoch": 6.885245901639344, "eval_accuracy": 0.8819264629725531, "eval_auc_class_0": 0.9495999423161078, "eval_auc_class_1": 0.9589146175353072, "eval_auc_class_2": 0.9645625643801177, "eval_auc_overall": 0.9576923747438442, "eval_loss": 0.3063272535800934, "eval_runtime": 210.6714, "eval_samples_per_second": 9.166, "eval_steps_per_second": 0.076, "step": 105 }, { "epoch": 6.885245901639344, "step": 105, "total_flos": 4.1785312376666235e+18, "train_loss": 0.4226800623394194, "train_runtime": 11867.0985, "train_samples_per_second": 4.564, "train_steps_per_second": 0.009 }, { "epoch": 6.885245901639344, "step": 105, "train_accuracy": 0.8871655680496316, "train_auc_class_0": 0.9601424899214485, "train_auc_class_1": 0.968345723853034, "train_auc_class_2": 0.9621813081120092, "train_auc_overall": 0.9635565072954971, "train_loss": 0.2833240032196045, "train_runtime": 718.0615, "train_samples_per_second": 10.775, "train_steps_per_second": 0.085 } ], "logging_steps": 10, "max_steps": 105, "num_input_tokens_seen": 0, "num_train_epochs": 7, "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": 4.1785312376666235e+18, "train_batch_size": 128, "trial_name": null, "trial_params": null }