| { | |
| "best_metric": 0.9818782216494845, | |
| "best_model_checkpoint": "../models/PoliticES/ideology_binary/BETO/450_SEED_11/run-1/checkpoint-3180", | |
| "epoch": 14.99647473560517, | |
| "global_step": 3180, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 2.7172184226965135e-05, | |
| "loss": 0.4738, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7511111111111111, | |
| "eval_jaccard": 0.5818173817381738, | |
| "eval_loss": 0.56746506690979, | |
| "eval_macro_f1": 0.716113914924297, | |
| "eval_macro_precision": 0.8270380109968922, | |
| "eval_macro_recall": 0.7193362193362193, | |
| "eval_micro_f1": 0.7511111111111111, | |
| "eval_micro_precision": 0.7511111111111111, | |
| "eval_micro_recall": 0.7511111111111111, | |
| "eval_runtime": 3.8883, | |
| "eval_samples_per_second": 387.83, | |
| "eval_steps_per_second": 12.345, | |
| "eval_weighted_f1": 0.7280749819754866, | |
| "eval_weighted_precision": 0.8113889552952426, | |
| "eval_weighted_recall": 0.7511111111111111, | |
| "step": 212 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "learning_rate": 2.7172184226965135e-05, | |
| "loss": 0.3003, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9422222222222222, | |
| "eval_jaccard": 0.8905400934915497, | |
| "eval_loss": 0.2769545614719391, | |
| "eval_macro_f1": 0.9410294147295417, | |
| "eval_macro_precision": 0.9443722592171513, | |
| "eval_macro_recall": 0.9386724386724387, | |
| "eval_micro_f1": 0.9422222222222222, | |
| "eval_micro_precision": 0.9422222222222222, | |
| "eval_micro_recall": 0.9422222222222222, | |
| "eval_runtime": 3.8935, | |
| "eval_samples_per_second": 387.316, | |
| "eval_steps_per_second": 12.328, | |
| "eval_weighted_f1": 0.9420358460514909, | |
| "eval_weighted_precision": 0.9428033133019327, | |
| "eval_weighted_recall": 0.9422222222222222, | |
| "step": 424 | |
| }, | |
| { | |
| "epoch": 2.83, | |
| "learning_rate": 2.7172184226965135e-05, | |
| "loss": 0.1564, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9333333333333333, | |
| "eval_jaccard": 0.8754464285714286, | |
| "eval_loss": 0.3916676938533783, | |
| "eval_macro_f1": 0.9329945005856777, | |
| "eval_macro_precision": 0.9322134387351779, | |
| "eval_macro_recall": 0.9383116883116883, | |
| "eval_micro_f1": 0.9333333333333333, | |
| "eval_micro_precision": 0.9333333333333333, | |
| "eval_micro_recall": 0.9333333333333333, | |
| "eval_runtime": 3.9044, | |
| "eval_samples_per_second": 386.23, | |
| "eval_steps_per_second": 12.294, | |
| "eval_weighted_f1": 0.9335662808473466, | |
| "eval_weighted_precision": 0.9382608695652174, | |
| "eval_weighted_recall": 0.9333333333333333, | |
| "step": 636 | |
| }, | |
| { | |
| "epoch": 3.77, | |
| "learning_rate": 2.7172184226965135e-05, | |
| "loss": 0.0897, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9422222222222222, | |
| "eval_jaccard": 0.8902276505022765, | |
| "eval_loss": 0.4636266231536865, | |
| "eval_macro_f1": 0.9407042510490785, | |
| "eval_macro_precision": 0.9481481481481482, | |
| "eval_macro_recall": 0.9365079365079365, | |
| "eval_micro_f1": 0.9422222222222222, | |
| "eval_micro_precision": 0.9422222222222222, | |
| "eval_micro_recall": 0.9422222222222222, | |
| "eval_runtime": 3.8881, | |
| "eval_samples_per_second": 387.847, | |
| "eval_steps_per_second": 12.345, | |
| "eval_weighted_f1": 0.9418427294289362, | |
| "eval_weighted_precision": 0.9445925925925925, | |
| "eval_weighted_recall": 0.9422222222222222, | |
| "step": 848 | |
| }, | |
| { | |
| "epoch": 4.71, | |
| "learning_rate": 2.7172184226965135e-05, | |
| "loss": 0.0562, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9422222222222222, | |
| "eval_jaccard": 0.8902276505022765, | |
| "eval_loss": 0.5073122978210449, | |
| "eval_macro_f1": 0.9407042510490785, | |
| "eval_macro_precision": 0.9481481481481482, | |
| "eval_macro_recall": 0.9365079365079365, | |
| "eval_micro_f1": 0.9422222222222222, | |
| "eval_micro_precision": 0.9422222222222222, | |
| "eval_micro_recall": 0.9422222222222222, | |
| "eval_runtime": 3.9083, | |
| "eval_samples_per_second": 385.848, | |
| "eval_steps_per_second": 12.282, | |
| "eval_weighted_f1": 0.9418427294289362, | |
| "eval_weighted_precision": 0.9445925925925925, | |
| "eval_weighted_recall": 0.9422222222222222, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 5.66, | |
| "learning_rate": 2.7172184226965135e-05, | |
| "loss": 0.0804, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.96, | |
| "eval_jaccard": 0.9233056148355289, | |
| "eval_loss": 0.6428035497665405, | |
| "eval_macro_f1": 0.9596814335490294, | |
| "eval_macro_precision": 0.9580624702711273, | |
| "eval_macro_recall": 0.9632034632034632, | |
| "eval_micro_f1": 0.96, | |
| "eval_micro_precision": 0.96, | |
| "eval_micro_recall": 0.96, | |
| "eval_runtime": 3.9157, | |
| "eval_samples_per_second": 385.112, | |
| "eval_steps_per_second": 12.258, | |
| "eval_weighted_f1": 0.9601114982578397, | |
| "eval_weighted_precision": 0.9620865704772475, | |
| "eval_weighted_recall": 0.96, | |
| "step": 1272 | |
| }, | |
| { | |
| "epoch": 6.6, | |
| "learning_rate": 2.7172184226965135e-05, | |
| "loss": 0.0179, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.9555555555555556, | |
| "eval_jaccard": 0.9148573109738158, | |
| "eval_loss": 0.630973756313324, | |
| "eval_macro_f1": 0.9548047565482886, | |
| "eval_macro_precision": 0.9559439432989691, | |
| "eval_macro_recall": 0.9538239538239539, | |
| "eval_micro_f1": 0.9555555555555556, | |
| "eval_micro_precision": 0.9555555555555556, | |
| "eval_micro_recall": 0.9555555555555556, | |
| "eval_runtime": 3.9129, | |
| "eval_samples_per_second": 385.393, | |
| "eval_steps_per_second": 12.267, | |
| "eval_weighted_f1": 0.9555037763136751, | |
| "eval_weighted_precision": 0.9556056701030928, | |
| "eval_weighted_recall": 0.9555555555555556, | |
| "step": 1484 | |
| }, | |
| { | |
| "epoch": 7.55, | |
| "learning_rate": 2.7172184226965135e-05, | |
| "loss": 0.0454, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.9644444444444444, | |
| "eval_jaccard": 0.9312917227959886, | |
| "eval_loss": 0.5245578289031982, | |
| "eval_macro_f1": 0.9638438052386309, | |
| "eval_macro_precision": 0.9650048324742269, | |
| "eval_macro_recall": 0.9628427128427128, | |
| "eval_micro_f1": 0.9644444444444444, | |
| "eval_micro_precision": 0.9644444444444444, | |
| "eval_micro_recall": 0.9644444444444444, | |
| "eval_runtime": 3.9143, | |
| "eval_samples_per_second": 385.257, | |
| "eval_steps_per_second": 12.263, | |
| "eval_weighted_f1": 0.9644030210509401, | |
| "eval_weighted_precision": 0.9645167525773195, | |
| "eval_weighted_recall": 0.9644444444444444, | |
| "step": 1696 | |
| }, | |
| { | |
| "epoch": 8.49, | |
| "learning_rate": 2.7172184226965135e-05, | |
| "loss": 0.0276, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.9511111111111111, | |
| "eval_jaccard": 0.9064495782911624, | |
| "eval_loss": 0.645963728427887, | |
| "eval_macro_f1": 0.9499686672461541, | |
| "eval_macro_precision": 0.9552958483164433, | |
| "eval_macro_recall": 0.9466089466089466, | |
| "eval_micro_f1": 0.9511111111111111, | |
| "eval_micro_precision": 0.9511111111111111, | |
| "eval_micro_recall": 0.9511111111111111, | |
| "eval_runtime": 3.8963, | |
| "eval_samples_per_second": 387.03, | |
| "eval_steps_per_second": 12.319, | |
| "eval_weighted_f1": 0.9508759020800905, | |
| "eval_weighted_precision": 0.9525400457665903, | |
| "eval_weighted_recall": 0.9511111111111111, | |
| "step": 1908 | |
| }, | |
| { | |
| "epoch": 9.43, | |
| "learning_rate": 2.7172184226965135e-05, | |
| "loss": 0.0244, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.9688888888888889, | |
| "eval_jaccard": 0.9395812863729122, | |
| "eval_loss": 0.5576620697975159, | |
| "eval_macro_f1": 0.9683257918552036, | |
| "eval_macro_precision": 0.970203488372093, | |
| "eval_macro_recall": 0.9668109668109668, | |
| "eval_micro_f1": 0.9688888888888889, | |
| "eval_micro_precision": 0.9688888888888889, | |
| "eval_micro_recall": 0.9688888888888889, | |
| "eval_runtime": 3.8888, | |
| "eval_samples_per_second": 387.781, | |
| "eval_steps_per_second": 12.343, | |
| "eval_weighted_f1": 0.9688325791855203, | |
| "eval_weighted_precision": 0.9691279069767442, | |
| "eval_weighted_recall": 0.9688888888888889, | |
| "step": 2120 | |
| }, | |
| { | |
| "epoch": 10.38, | |
| "learning_rate": 2.7172184226965135e-05, | |
| "loss": 0.0144, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.9688888888888889, | |
| "eval_jaccard": 0.9397094904794161, | |
| "eval_loss": 0.6369400024414062, | |
| "eval_macro_f1": 0.9684678371939377, | |
| "eval_macro_precision": 0.968, | |
| "eval_macro_recall": 0.968975468975469, | |
| "eval_micro_f1": 0.9688888888888889, | |
| "eval_micro_precision": 0.9688888888888889, | |
| "eval_micro_recall": 0.9688888888888889, | |
| "eval_runtime": 3.8878, | |
| "eval_samples_per_second": 387.876, | |
| "eval_steps_per_second": 12.346, | |
| "eval_weighted_f1": 0.9689050831848484, | |
| "eval_weighted_precision": 0.9689599999999999, | |
| "eval_weighted_recall": 0.9688888888888889, | |
| "step": 2332 | |
| }, | |
| { | |
| "epoch": 11.32, | |
| "learning_rate": 2.7172184226965135e-05, | |
| "loss": 0.0113, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.9688888888888889, | |
| "eval_jaccard": 0.9395812863729122, | |
| "eval_loss": 0.6348240971565247, | |
| "eval_macro_f1": 0.9683257918552036, | |
| "eval_macro_precision": 0.970203488372093, | |
| "eval_macro_recall": 0.9668109668109668, | |
| "eval_micro_f1": 0.9688888888888889, | |
| "eval_micro_precision": 0.9688888888888889, | |
| "eval_micro_recall": 0.9688888888888889, | |
| "eval_runtime": 3.8915, | |
| "eval_samples_per_second": 387.511, | |
| "eval_steps_per_second": 12.335, | |
| "eval_weighted_f1": 0.9688325791855203, | |
| "eval_weighted_precision": 0.9691279069767442, | |
| "eval_weighted_recall": 0.9688888888888889, | |
| "step": 2544 | |
| }, | |
| { | |
| "epoch": 12.26, | |
| "learning_rate": 2.7172184226965135e-05, | |
| "loss": 0.017, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.9333333333333333, | |
| "eval_jaccard": 0.8737588652482269, | |
| "eval_loss": 0.6715770363807678, | |
| "eval_macro_f1": 0.9309265058021734, | |
| "eval_macro_precision": 0.9468085106382979, | |
| "eval_macro_recall": 0.9242424242424243, | |
| "eval_micro_f1": 0.9333333333333333, | |
| "eval_micro_precision": 0.9333333333333333, | |
| "eval_micro_recall": 0.9333333333333333, | |
| "eval_runtime": 3.8914, | |
| "eval_samples_per_second": 387.517, | |
| "eval_steps_per_second": 12.335, | |
| "eval_weighted_f1": 0.9324737520722048, | |
| "eval_weighted_precision": 0.9404255319148935, | |
| "eval_weighted_recall": 0.9333333333333333, | |
| "step": 2756 | |
| }, | |
| { | |
| "epoch": 13.21, | |
| "learning_rate": 2.7172184226965135e-05, | |
| "loss": 0.0162, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.9555555555555556, | |
| "eval_jaccard": 0.9146445987882369, | |
| "eval_loss": 0.6302870512008667, | |
| "eval_macro_f1": 0.954578488372093, | |
| "eval_macro_precision": 0.9589442815249267, | |
| "eval_macro_recall": 0.9516594516594516, | |
| "eval_micro_f1": 0.9555555555555556, | |
| "eval_micro_precision": 0.9555555555555556, | |
| "eval_micro_recall": 0.9555555555555556, | |
| "eval_runtime": 3.8937, | |
| "eval_samples_per_second": 387.288, | |
| "eval_steps_per_second": 12.327, | |
| "eval_weighted_f1": 0.9553779069767442, | |
| "eval_weighted_precision": 0.9565982404692083, | |
| "eval_weighted_recall": 0.9555555555555556, | |
| "step": 2968 | |
| }, | |
| { | |
| "epoch": 14.15, | |
| "learning_rate": 2.7172184226965135e-05, | |
| "loss": 0.0134, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.9822222222222222, | |
| "eval_jaccard": 0.964991452991453, | |
| "eval_loss": 0.6276100277900696, | |
| "eval_macro_f1": 0.9818782216494845, | |
| "eval_macro_precision": 0.9846153846153847, | |
| "eval_macro_recall": 0.9797979797979798, | |
| "eval_micro_f1": 0.9822222222222222, | |
| "eval_micro_precision": 0.9822222222222222, | |
| "eval_micro_recall": 0.9822222222222222, | |
| "eval_runtime": 3.8919, | |
| "eval_samples_per_second": 387.474, | |
| "eval_steps_per_second": 12.333, | |
| "eval_weighted_f1": 0.9821778350515463, | |
| "eval_weighted_precision": 0.9827692307692307, | |
| "eval_weighted_recall": 0.9822222222222222, | |
| "step": 3180 | |
| } | |
| ], | |
| "max_steps": 4240, | |
| "num_train_epochs": 20, | |
| "total_flos": 4.724653668328427e+16, | |
| "trial_name": null, | |
| "trial_params": { | |
| "attention_probs_dropout_prob": 0.12254664017551167, | |
| "gradient_accumulation_steps": 4, | |
| "hidden_dropout_prob": 0.09767216663102402, | |
| "learning_rate": 2.7172184226965135e-05, | |
| "lr_scheduler_type": "constant", | |
| "num_train_epochs": 20, | |
| "per_device_train_batch_size": 16, | |
| "weight_decay": 0.004770259527907836 | |
| } | |
| } | |