| { |
| "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 |
| } |
| } |
|
|
