{ "best_metric": 0.7546452702702704, "best_model_checkpoint": "./output/neuralmind/bert-base-portuguese-cased-finetune-wiki_section_disease/seed42-seq512-lr5e-05-epoch5-bs4-ts1.0-tssp1.0-cl0.5-2024-05-03_16:56:29/checkpoint-13420", "epoch": 5.0, "global_step": 13425, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 1.0, "learning_rate": 4.000372439478585e-05, "loss": 0.9152, "step": 2684 }, { "epoch": 1.0, "eval_EOP_f1": 0.7217790669567319, "eval_EOP_number": 2446, "eval_EOP_precision": 0.882909520993495, "eval_EOP_recall": 0.6103843008994276, "eval_da_EOP_f1": 0.7340590979782271, "eval_da_EOP_number": 2704, "eval_da_EOP_precision": 0.9193099610461881, "eval_da_EOP_recall": 0.6109467455621301, "eval_da_overall_accuracy": 0.9443618109138235, "eval_da_overall_f1": 0.7340590979782271, "eval_da_overall_precision": 0.9193099610461881, "eval_da_overall_recall": 0.6109467455621301, "eval_loss": NaN, "eval_overall_accuracy": 0.9441993503660251, "eval_overall_f1": 0.7217790669567319, "eval_overall_precision": 0.882909520993495, "eval_overall_recall": 0.6103843008994276, "eval_runtime": 115.3406, "eval_samples_per_second": 13.204, "eval_steps_per_second": 3.303, "step": 2684 }, { "epoch": 2.0, "learning_rate": 3.0007448789571697e-05, "loss": 0.6588, "step": 5368 }, { "epoch": 2.0, "eval_EOP_f1": 0.7412026726057906, "eval_EOP_number": 2446, "eval_EOP_precision": 0.8140900195694716, "eval_EOP_recall": 0.6802943581357318, "eval_da_EOP_f1": 0.7685429790692948, "eval_da_EOP_number": 2704, "eval_da_EOP_precision": 0.8529544429409112, "eval_da_EOP_recall": 0.6993343195266272, "eval_da_overall_accuracy": 0.94705772985033, "eval_da_overall_f1": 0.7685429790692948, "eval_da_overall_precision": 0.8529544429409112, "eval_da_overall_recall": 0.6993343195266272, "eval_loss": NaN, "eval_overall_accuracy": 0.9436660687448489, "eval_overall_f1": 0.7412026726057906, "eval_overall_precision": 0.8140900195694716, "eval_overall_recall": 0.6802943581357318, "eval_runtime": 115.0795, "eval_samples_per_second": 13.234, "eval_steps_per_second": 3.311, "step": 5368 }, { "epoch": 3.0, "learning_rate": 2.0011173184357543e-05, "loss": 0.4949, "step": 8052 }, { "epoch": 3.0, "eval_EOP_f1": 0.7511530858774434, "eval_EOP_number": 2446, "eval_EOP_precision": 0.8115804461319411, "eval_EOP_recall": 0.6991005723630417, "eval_da_EOP_f1": 0.7725617159044921, "eval_da_EOP_number": 2704, "eval_da_EOP_precision": 0.8529937444146559, "eval_da_EOP_recall": 0.705991124260355, "eval_da_overall_accuracy": 0.9477549502649437, "eval_da_overall_f1": 0.7725617159044921, "eval_da_overall_precision": 0.8529937444146559, "eval_da_overall_recall": 0.705991124260355, "eval_loss": NaN, "eval_overall_accuracy": 0.9450719930188588, "eval_overall_f1": 0.7511530858774434, "eval_overall_precision": 0.8115804461319411, "eval_overall_recall": 0.6991005723630417, "eval_runtime": 114.897, "eval_samples_per_second": 13.255, "eval_steps_per_second": 3.316, "step": 8052 }, { "epoch": 4.0, "learning_rate": 1.001489757914339e-05, "loss": 0.3741, "step": 10736 }, { "epoch": 4.0, "eval_EOP_f1": 0.752757949383517, "eval_EOP_number": 2446, "eval_EOP_precision": 0.799265043638034, "eval_EOP_recall": 0.7113654946852003, "eval_da_EOP_f1": 0.7717067355454003, "eval_da_EOP_number": 2704, "eval_da_EOP_precision": 0.8338342636324603, "eval_da_EOP_recall": 0.7181952662721893, "eval_da_overall_accuracy": 0.9465929162405875, "eval_da_overall_f1": 0.7717067355454003, "eval_da_overall_precision": 0.8338342636324603, "eval_da_overall_recall": 0.7181952662721893, "eval_loss": NaN, "eval_overall_accuracy": 0.9445871915450623, "eval_overall_f1": 0.752757949383517, "eval_overall_precision": 0.799265043638034, "eval_overall_recall": 0.7113654946852003, "eval_runtime": 114.8732, "eval_samples_per_second": 13.258, "eval_steps_per_second": 3.317, "step": 10736 }, { "epoch": 5.0, "learning_rate": 1.86219739292365e-08, "loss": 0.2844, "step": 13420 }, { "epoch": 5.0, "eval_EOP_f1": 0.7546452702702704, "eval_EOP_number": 2446, "eval_EOP_precision": 0.780349344978166, "eval_EOP_recall": 0.7305805396565822, "eval_da_EOP_f1": 0.7679782903663502, "eval_da_EOP_number": 2704, "eval_da_EOP_precision": 0.8069246435845214, "eval_da_EOP_recall": 0.7326183431952663, "eval_da_overall_accuracy": 0.9443618109138235, "eval_da_overall_f1": 0.7679782903663502, "eval_da_overall_precision": 0.8069246435845214, "eval_da_overall_recall": 0.7326183431952663, "eval_loss": NaN, "eval_overall_accuracy": 0.9436660687448489, "eval_overall_f1": 0.7546452702702704, "eval_overall_precision": 0.780349344978166, "eval_overall_recall": 0.7305805396565822, "eval_runtime": 114.7975, "eval_samples_per_second": 13.267, "eval_steps_per_second": 3.319, "step": 13420 }, { "epoch": 5.0, "step": 13425, "total_flos": 2.825625731920896e+16, "train_loss": 0.545391144610428, "train_runtime": 12631.9735, "train_samples_per_second": 4.251, "train_steps_per_second": 1.063 } ], "max_steps": 13425, "num_train_epochs": 5, "total_flos": 2.825625731920896e+16, "trial_name": null, "trial_params": null }