| { | |
| "best_metric": 0.8328207869559483, | |
| "best_model_checkpoint": "./CARES/checkpoints/roberta-stratified/run-9/checkpoint-2272", | |
| "epoch": 32.0, | |
| "global_step": 2272, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.29745519161224365, | |
| "eval_macro_f1": 0.0, | |
| "eval_macro_precision": 0.0, | |
| "eval_macro_recall": 0.0, | |
| "eval_micro_f1": 0.0, | |
| "eval_micro_precision": 0.0, | |
| "eval_micro_recall": 0.0, | |
| "eval_runtime": 2.6698, | |
| "eval_samples_per_second": 361.828, | |
| "eval_steps_per_second": 11.611, | |
| "step": 142 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.2143602967262268, | |
| "eval_macro_f1": 0.20760247776727883, | |
| "eval_macro_precision": 0.2427021619684663, | |
| "eval_macro_recall": 0.19452630677199997, | |
| "eval_micro_f1": 0.5696619950535862, | |
| "eval_micro_precision": 0.7601760176017601, | |
| "eval_micro_recall": 0.45550428477257743, | |
| "eval_runtime": 2.6814, | |
| "eval_samples_per_second": 360.254, | |
| "eval_steps_per_second": 11.561, | |
| "step": 284 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 0.15995128452777863, | |
| "eval_macro_f1": 0.32385167310781526, | |
| "eval_macro_precision": 0.42763364413607774, | |
| "eval_macro_recall": 0.2931228227099484, | |
| "eval_micro_f1": 0.6900908014212397, | |
| "eval_micro_precision": 0.860236220472441, | |
| "eval_micro_recall": 0.5761371127224786, | |
| "eval_runtime": 2.674, | |
| "eval_samples_per_second": 361.255, | |
| "eval_steps_per_second": 11.593, | |
| "step": 426 | |
| }, | |
| { | |
| "epoch": 7.04, | |
| "learning_rate": 4.40741152818786e-05, | |
| "loss": 0.2773, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 0.13520723581314087, | |
| "eval_macro_f1": 0.4425371105794407, | |
| "eval_macro_precision": 0.49169850541474464, | |
| "eval_macro_recall": 0.41330740987880255, | |
| "eval_micro_f1": 0.7547309833024118, | |
| "eval_micro_precision": 0.8633276740237691, | |
| "eval_micro_recall": 0.6704021094264997, | |
| "eval_runtime": 2.6984, | |
| "eval_samples_per_second": 357.994, | |
| "eval_steps_per_second": 11.488, | |
| "step": 568 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 0.10962910205125809, | |
| "eval_macro_f1": 0.5655053778813368, | |
| "eval_macro_precision": 0.6696587937430376, | |
| "eval_macro_recall": 0.5338063461136978, | |
| "eval_micro_f1": 0.8189806678383128, | |
| "eval_micro_precision": 0.8772590361445783, | |
| "eval_micro_recall": 0.7679630850362558, | |
| "eval_runtime": 2.6744, | |
| "eval_samples_per_second": 361.208, | |
| "eval_steps_per_second": 11.592, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 0.0956677794456482, | |
| "eval_macro_f1": 0.6602463081601572, | |
| "eval_macro_precision": 0.799820247637967, | |
| "eval_macro_recall": 0.5970945139878616, | |
| "eval_micro_f1": 0.8437173686042465, | |
| "eval_micro_precision": 0.8938053097345132, | |
| "eval_micro_recall": 0.7989452867501649, | |
| "eval_runtime": 2.6891, | |
| "eval_samples_per_second": 359.223, | |
| "eval_steps_per_second": 11.528, | |
| "step": 852 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_loss": 0.09418635815382004, | |
| "eval_macro_f1": 0.709171253471717, | |
| "eval_macro_precision": 0.8508463365856882, | |
| "eval_macro_recall": 0.6566230286073916, | |
| "eval_micro_f1": 0.8551865799383773, | |
| "eval_micro_precision": 0.8896011396011396, | |
| "eval_micro_recall": 0.8233355306526038, | |
| "eval_runtime": 2.6733, | |
| "eval_samples_per_second": 361.353, | |
| "eval_steps_per_second": 11.596, | |
| "step": 994 | |
| }, | |
| { | |
| "epoch": 14.08, | |
| "learning_rate": 4.073516715446356e-05, | |
| "loss": 0.068, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_loss": 0.09525582939386368, | |
| "eval_macro_f1": 0.776452344525862, | |
| "eval_macro_precision": 0.8422867152488994, | |
| "eval_macro_recall": 0.7394950574942071, | |
| "eval_micro_f1": 0.8662207357859532, | |
| "eval_micro_precision": 0.8791581805838425, | |
| "eval_micro_recall": 0.8536585365853658, | |
| "eval_runtime": 2.6899, | |
| "eval_samples_per_second": 359.119, | |
| "eval_steps_per_second": 11.525, | |
| "step": 1136 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_loss": 0.09120669960975647, | |
| "eval_macro_f1": 0.7799395687308482, | |
| "eval_macro_precision": 0.8259713713716451, | |
| "eval_macro_recall": 0.7560566435810081, | |
| "eval_micro_f1": 0.8660743665679499, | |
| "eval_micro_precision": 0.8646517739816032, | |
| "eval_micro_recall": 0.8675016479894528, | |
| "eval_runtime": 2.6882, | |
| "eval_samples_per_second": 359.349, | |
| "eval_steps_per_second": 11.532, | |
| "step": 1278 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_loss": 0.09322977066040039, | |
| "eval_macro_f1": 0.7717526983031062, | |
| "eval_macro_precision": 0.8213810289257493, | |
| "eval_macro_recall": 0.747806253729357, | |
| "eval_micro_f1": 0.8603205757278378, | |
| "eval_micro_precision": 0.8538961038961039, | |
| "eval_micro_recall": 0.8668424522083059, | |
| "eval_runtime": 2.6911, | |
| "eval_samples_per_second": 358.966, | |
| "eval_steps_per_second": 11.52, | |
| "step": 1420 | |
| }, | |
| { | |
| "epoch": 21.13, | |
| "learning_rate": 3.739621902704851e-05, | |
| "loss": 0.0222, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_loss": 0.10442250967025757, | |
| "eval_macro_f1": 0.765921285849591, | |
| "eval_macro_precision": 0.7782427042161657, | |
| "eval_macro_recall": 0.774471020810195, | |
| "eval_micro_f1": 0.8516003879728419, | |
| "eval_micro_precision": 0.8356598984771574, | |
| "eval_micro_recall": 0.8681608437705999, | |
| "eval_runtime": 2.6921, | |
| "eval_samples_per_second": 358.824, | |
| "eval_steps_per_second": 11.515, | |
| "step": 1562 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_loss": 0.09634574502706528, | |
| "eval_macro_f1": 0.805432074935829, | |
| "eval_macro_precision": 0.9021531097855335, | |
| "eval_macro_recall": 0.7712696478949495, | |
| "eval_micro_f1": 0.8704318936877077, | |
| "eval_micro_precision": 0.8774279973208305, | |
| "eval_micro_recall": 0.8635464733025708, | |
| "eval_runtime": 2.6753, | |
| "eval_samples_per_second": 361.083, | |
| "eval_steps_per_second": 11.588, | |
| "step": 1704 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_loss": 0.09631907194852829, | |
| "eval_macro_f1": 0.8054621279402976, | |
| "eval_macro_precision": 0.8819456215653025, | |
| "eval_macro_recall": 0.782841919580731, | |
| "eval_micro_f1": 0.8722700198544011, | |
| "eval_micro_precision": 0.8757475083056478, | |
| "eval_micro_recall": 0.8688200395517469, | |
| "eval_runtime": 2.6758, | |
| "eval_samples_per_second": 361.016, | |
| "eval_steps_per_second": 11.585, | |
| "step": 1846 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_loss": 0.09827031195163727, | |
| "eval_macro_f1": 0.824449032097833, | |
| "eval_macro_precision": 0.8941626132550853, | |
| "eval_macro_recall": 0.7913713438345684, | |
| "eval_micro_f1": 0.8718459495351926, | |
| "eval_micro_precision": 0.8782608695652174, | |
| "eval_micro_recall": 0.8655240606460118, | |
| "eval_runtime": 2.6758, | |
| "eval_samples_per_second": 361.008, | |
| "eval_steps_per_second": 11.585, | |
| "step": 1988 | |
| }, | |
| { | |
| "epoch": 28.17, | |
| "learning_rate": 3.4057270899633464e-05, | |
| "loss": 0.011, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_loss": 0.10059890896081924, | |
| "eval_macro_f1": 0.8165877105799546, | |
| "eval_macro_precision": 0.884478008536637, | |
| "eval_macro_recall": 0.7957804401453603, | |
| "eval_micro_f1": 0.8693088765149033, | |
| "eval_micro_precision": 0.8639322916666666, | |
| "eval_micro_recall": 0.8747528015820699, | |
| "eval_runtime": 2.6759, | |
| "eval_samples_per_second": 361.006, | |
| "eval_steps_per_second": 11.585, | |
| "step": 2130 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_loss": 0.10830199718475342, | |
| "eval_macro_f1": 0.8328207869559483, | |
| "eval_macro_precision": 0.8739781063350807, | |
| "eval_macro_recall": 0.8212976019561394, | |
| "eval_micro_f1": 0.8676422764227643, | |
| "eval_micro_precision": 0.8562259306803595, | |
| "eval_micro_recall": 0.8793671720500988, | |
| "eval_runtime": 2.674, | |
| "eval_samples_per_second": 361.259, | |
| "eval_steps_per_second": 11.593, | |
| "step": 2272 | |
| } | |
| ], | |
| "max_steps": 7100, | |
| "num_train_epochs": 100, | |
| "total_flos": 1.673289313217472e+16, | |
| "trial_name": null, | |
| "trial_params": { | |
| "adam_epsilon": 1.724439344881123e-07, | |
| "learning_rate": 4.540969453284462e-05, | |
| "per_device_eval_batch_size": 32, | |
| "per_device_train_batch_size": 32, | |
| "seed": 324, | |
| "warmup_steps": 300, | |
| "weight_decay": 0.00598936569463419 | |
| } | |
| } | |