updated model

eval_results.txt

@@ -1,4 +0,0 @@
-eval_loss = 0.061167137617421395
-f1_score = 0.9648662235379545
-precision = 0.9612276346203236
-recall = 0.9685324638356235

model_args.json

@@ -1 +1 @@
-{"adafactor_beta1": null, "adafactor_clip_threshold": 1.0, "adafactor_decay_rate": -0.8, "adafactor_eps": [1e-30, 0.001], "adafactor_relative_step": true, "adafactor_scale_parameter": true, "adafactor_warmup_init": true, "adam_epsilon": 1e-08, "best_model_dir": "outputs/best_model", "cache_dir": "cache_dir/", "config": {}, "cosine_schedule_num_cycles": 0.5, "custom_layer_parameters": [], "custom_parameter_groups": [], "dataloader_num_workers": 0, "do_lower_case": false, "dynamic_quantize": false, "early_stopping_consider_epochs": false, "early_stopping_delta": 0, "early_stopping_metric": "eval_loss", "early_stopping_metric_minimize": true, "early_stopping_patience": 3, "encoding": null, "eval_batch_size": 8, "evaluate_during_training": true, "evaluate_during_training_silent": true, "evaluate_during_training_steps": 500, "evaluate_during_training_verbose": true, "evaluate_each_epoch": true, "fp16": true, "gradient_accumulation_steps": 1, "learning_rate": 4e-05, "local_rank": -1, "logging_steps": 50, "loss_type": null, "loss_args": {}, "manual_seed": null, "max_grad_norm": 1.0, "max_seq_length": 128, "model_name": "bert-large-cased", "model_type": "bert", "multiprocessing_chunksize": -1, "n_gpu": 1, "no_cache": false, "no_save": false, "not_saved_args": [], "num_train_epochs": 5, "optimizer": "AdamW", "output_dir": "../sent-tokenizer/bert-large-cased_1", "overwrite_output_dir": false, "polynomial_decay_schedule_lr_end": 1e-07, "polynomial_decay_schedule_power": 1.0, "process_count": 46, "quantized_model": false, "reprocess_input_data": true, "save_best_model": true, "save_eval_checkpoints": true, "save_model_every_epoch": true, "save_optimizer_and_scheduler": true, "save_steps": 2000, "scheduler": "linear_schedule_with_warmup", "silent": false, "skip_special_tokens": true, "tensorboard_dir": null, "thread_count": null, "tokenizer_name": null, "tokenizer_type": null, "train_batch_size": 8, "train_custom_parameters_only": false, "use_cached_eval_features": false, "use_early_stopping": false, "use_hf_datasets": false, "use_multiprocessing": true, "use_multiprocessing_for_evaluation": true, "wandb_kwargs": {"notes": "bert-base-cased. weight=[0.2,0.7,0.1]"}, "wandb_project": "sent-tokenization", "warmup_ratio": 0.06, "warmup_steps": 706, "weight_decay": 0.0, "model_class": "NERModel", "classification_report": false, "labels_list": [], "lazy_loading": false, "lazy_loading_start_line": 0, "onnx": false, "special_tokens_list": []}
+{"adafactor_beta1": null, "adafactor_clip_threshold": 1.0, "adafactor_decay_rate": -0.8, "adafactor_eps": [1e-30, 0.001], "adafactor_relative_step": true, "adafactor_scale_parameter": true, "adafactor_warmup_init": true, "adam_epsilon": 1e-08, "best_model_dir": "outputs/best_model", "cache_dir": "cache_dir/", "config": {}, "cosine_schedule_num_cycles": 0.5, "custom_layer_parameters": [], "custom_parameter_groups": [], "dataloader_num_workers": 0, "do_lower_case": false, "dynamic_quantize": false, "early_stopping_consider_epochs": false, "early_stopping_delta": 0, "early_stopping_metric": "eval_loss", "early_stopping_metric_minimize": true, "early_stopping_patience": 3, "encoding": null, "eval_batch_size": 8, "evaluate_during_training": true, "evaluate_during_training_silent": true, "evaluate_during_training_steps": 500, "evaluate_during_training_verbose": true, "evaluate_each_epoch": true, "fp16": true, "gradient_accumulation_steps": 1, "learning_rate": 4e-05, "local_rank": -1, "logging_steps": 50, "loss_type": null, "loss_args": {}, "manual_seed": null, "max_grad_norm": 1.0, "max_seq_length": 128, "model_name": "bert-large-cased", "model_type": "bert", "multiprocessing_chunksize": -1, "n_gpu": 1, "no_cache": false, "no_save": false, "not_saved_args": [], "num_train_epochs": 5, "optimizer": "AdamW", "output_dir": "../sent-tokenizer/bert-large-cased_1", "overwrite_output_dir": false, "polynomial_decay_schedule_lr_end": 1e-07, "polynomial_decay_schedule_power": 1.0, "process_count": 46, "quantized_model": false, "reprocess_input_data": true, "save_best_model": true, "save_eval_checkpoints": true, "save_model_every_epoch": true, "save_optimizer_and_scheduler": true, "save_steps": 2000, "scheduler": "linear_schedule_with_warmup", "silent": false, "skip_special_tokens": true, "tensorboard_dir": null, "thread_count": null, "tokenizer_name": null, "tokenizer_type": null, "train_batch_size": 8, "train_custom_parameters_only": false, "use_cached_eval_features": false, "use_early_stopping": false, "use_hf_datasets": false, "use_multiprocessing": true, "use_multiprocessing_for_evaluation": true, "wandb_kwargs": {"notes": "bert-base-cased. weight=[0.2,0.7,0.1]"}, "wandb_project": "sent-tokenization", "warmup_ratio": 0.06, "warmup_steps": 706, "weight_decay": 0.0, "model_class": "NERModel", "classification_report": false, "labels_list": ["B-sent", "B-split", "I-sent"], "lazy_loading": false, "lazy_loading_start_line": 0, "onnx": false, "special_tokens_list": []}

training_progress_scores.csv

@@ -1,29 +0,0 @@
-global_step,train_loss,eval_loss,precision,recall,f1_score
-500,0.007581932470202446,0.07520355188474497,0.8669141635729554,0.9306040766331116,0.8976307852314093
-1000,0.04216049239039421,0.06077631997306373,0.8607203574111414,0.9389453889880875,0.8981327920658102
-1500,0.0028899949975311756,0.057350501998572175,0.9241525813546989,0.9531696124351974,0.938436844474263
-2000,0.05302853509783745,0.06097400809681778,0.944389258066726,0.9512891511183569,0.9478266474721214
-2353,0.0032778969034552574,0.04456851387687447,0.902562676827972,0.9521172441753142,0.9266779446149197
-2500,0.018591150641441345,0.05848436432731241,0.9272330133650233,0.9557228993608157,0.9412624246028375
-3000,0.053025975823402405,0.05124464090028568,0.9310619194787139,0.9589145080178385,0.9447829819059518
-3500,0.0014473552582785487,0.05313493127800351,0.9422447408710657,0.9608898549974554,0.951475964330862
-4000,0.07083650678396225,0.04853879902211225,0.9446397337273083,0.9596735933200493,0.9520973201028656
-4500,0.009007592685520649,0.041883065983694186,0.9336678702175607,0.9613729092806804,0.9473178692551573
-4706,0.0003281784593127668,0.04338375121308074,0.938311797563766,0.9621319945828912,0.9500726146822203
-5000,0.010362996719777584,0.043102149878265474,0.9322141538050154,0.9623045139697574,0.9470203735144312
-5500,0.030205219984054565,0.0455509968784568,0.9445532518403458,0.9651424578837047,0.9547368645604455
-6000,0.0005797464400529861,0.04463238641486926,0.9482472199620288,0.9650561981902717,0.956577872969865
-6500,0.0075051793828606606,0.04087219411031742,0.9359086182025562,0.962606422896773,0.949069802053877
-7000,0.0074796611443161964,0.04322933663870992,0.9528950619913936,0.9646162737537631,0.9587198436240806
-7059,0.00048343068920075893,0.04181959683803667,0.9496054975820819,0.9655047485961235,0.957489125274274
-7500,0.0008980808779597282,0.04782667562439947,0.9568521193575684,0.9656341381362731,0.9612230703840771
-8000,0.019552884623408318,0.05795671805083763,0.9563792750358386,0.9667986439976193,0.9615607345541586
-8500,0.0004622614069376141,0.05230412267701325,0.9540909323203244,0.9662465819596477,0.9601302847836801
-9000,0.00027256805333308876,0.04744021896397177,0.9521660803208756,0.9665312389479768,0.9592948841430265
-9412,8.708691893843934e-05,0.05034951700891713,0.9571483359611369,0.9687567390385494,0.9629175526442143
-9500,0.0001998538209591061,0.055368390772068496,0.9578800058025924,0.968290936694011,0.9630573357698676
-10000,0.01602429524064064,0.05962661065816631,0.9575058012558012,0.9681356692458315,0.962791395912415
-10500,0.0036623200867325068,0.06314988210496593,0.962033146892801,0.9658670393085423,0.9639462809917354
-11000,0.0017030838644132018,0.06320890702416958,0.9613907886392075,0.9676353630239198,0.9645029684752656
-11500,0.008592898026108742,0.06004220840402733,0.9605725873383929,0.9683944483261306,0.9644676592125497
-11765,6.884566391818225e-05,0.061167137617421395,0.9612276346203236,0.9685324638356235,0.9648662235379545