retro-reader / results /intensive_eval_results.txt
jinmang2's picture
mv koelectra-small results
fca8219
raw
history blame
6.71 kB
***** Eval results *****
2021-12-13 21:36:24.587572epoch = 1.0
eval_HasAns_exact = 78.35122965015587
eval_HasAns_f1 = 85.13968688137153
eval_HasAns_total = 5774
eval_NoAns_exact = 82.26532732940768
eval_NoAns_f1 = 82.26532732940768
eval_NoAns_total = 5774
eval_best_exact = 83.36508486317977
eval_best_exact_thresh = -4.265625
eval_best_f1 = 86.50049020991854
eval_best_f1_thresh = -3.76171875
eval_exact = 80.30827848978178
eval_f1 = 83.70250710538936
eval_runtime = 116.2823
eval_samples_per_second = 119.743
eval_steps_per_second = 0.937
eval_total = 11548
***** Eval results *****
2021-12-13 23:54:17.975036epoch = 2.0
eval_HasAns_exact = 81.65916175961205
eval_HasAns_f1 = 87.7141782584969
eval_HasAns_total = 5774
eval_NoAns_exact = 79.14790439903014
eval_NoAns_f1 = 79.14790439903014
eval_NoAns_total = 5774
eval_best_exact = 85.417388292345
eval_best_exact_thresh = -4.91015625
eval_best_f1 = 88.2525034462934
eval_best_f1_thresh = -4.91015625
eval_exact = 80.4035330793211
eval_f1 = 83.4310413287634
eval_runtime = 114.7091
eval_samples_per_second = 121.385
eval_steps_per_second = 0.95
eval_total = 11548
***** Eval results *****
2021-12-14 02:12:16.445998epoch = 3.0
eval_HasAns_exact = 82.40387945964669
eval_HasAns_f1 = 88.17376210385724
eval_HasAns_total = 5774
eval_NoAns_exact = 82.04018011776931
eval_NoAns_f1 = 82.04018011776931
eval_NoAns_total = 5774
eval_best_exact = 85.93695878074125
eval_best_exact_thresh = -5.8046875
eval_best_f1 = 88.652083582146
eval_best_f1_thresh = -4.640625
eval_exact = 82.222029788708
eval_f1 = 85.10697111081303
eval_runtime = 115.8495
eval_samples_per_second = 120.19
eval_steps_per_second = 0.941
eval_total = 11548
***** Eval results *****
2021-12-14 04:30:40.843577epoch = 4.0
eval_HasAns_exact = 82.92344994804296
eval_HasAns_f1 = 88.76361604970366
eval_HasAns_total = 5774
eval_NoAns_exact = 75.68410114305507
eval_NoAns_f1 = 75.68410114305507
eval_NoAns_total = 5774
eval_best_exact = 86.31797713889851
eval_best_exact_thresh = -7.69921875
eval_best_f1 = 89.0138490084362
eval_best_f1_thresh = -7.4140625
eval_exact = 79.30377554554902
eval_f1 = 82.22385859637924
eval_runtime = 116.6262
eval_samples_per_second = 119.39
eval_steps_per_second = 0.935
eval_total = 11548
***** Eval results *****
2021-12-14 06:49:11.744655epoch = 5.0
eval_HasAns_exact = 83.04468306200208
eval_HasAns_f1 = 88.60010543614725
eval_HasAns_total = 5774
eval_NoAns_exact = 84.48216141323172
eval_NoAns_f1 = 84.48216141323172
eval_NoAns_total = 5774
eval_best_exact = 87.20990647731209
eval_best_exact_thresh = -4.83984375
eval_best_f1 = 89.8436215977173
eval_best_f1_thresh = -4.83984375
eval_exact = 83.76342223761691
eval_f1 = 86.5411334246892
eval_runtime = 117.7564
eval_samples_per_second = 118.244
eval_steps_per_second = 0.926
eval_total = 11548
***** Eval results *****
2021-12-14 09:07:43.312763epoch = 6.0
eval_HasAns_exact = 83.46033945271908
eval_HasAns_f1 = 89.05328761170561
eval_HasAns_total = 5774
eval_NoAns_exact = 82.05749913404918
eval_NoAns_f1 = 82.05749913404918
eval_NoAns_total = 5774
eval_best_exact = 87.53030827848978
eval_best_exact_thresh = -6.7421875
eval_best_f1 = 90.18823628334343
eval_best_f1_thresh = -5.328125
eval_exact = 82.75891929338414
eval_f1 = 85.55539337287718
eval_runtime = 115.8934
eval_samples_per_second = 120.145
eval_steps_per_second = 0.941
eval_total = 11548
***** Eval results *****
2021-12-14 11:26:17.714938epoch = 7.0
eval_HasAns_exact = 83.47765846899897
eval_HasAns_f1 = 88.97497847186122
eval_HasAns_total = 5774
eval_NoAns_exact = 83.28714928992034
eval_NoAns_f1 = 83.28714928992034
eval_NoAns_total = 5774
eval_best_exact = 87.56494631104954
eval_best_exact_thresh = -5.70703125
eval_best_f1 = 90.1712271871601
eval_best_f1_thresh = -5.6875
eval_exact = 83.38240387945964
eval_f1 = 86.13106388089054
eval_runtime = 118.9527
eval_samples_per_second = 117.055
eval_steps_per_second = 0.916
eval_total = 11548
***** Eval results *****
2021-12-14 13:22:43.923006epoch = 6.0
eval_HasAns_exact = 83.46033945271908
eval_HasAns_f1 = 89.05328761170561
eval_HasAns_total = 5774
eval_NoAns_exact = 82.05749913404918
eval_NoAns_f1 = 82.05749913404918
eval_NoAns_total = 5774
eval_best_exact = 87.53030827848978
eval_best_exact_thresh = -6.7421875
eval_best_f1 = 90.18823628334343
eval_best_f1_thresh = -5.328125
eval_exact = 82.75891929338414
eval_f1 = 85.55539337287718
eval_runtime = 118.456
eval_samples_per_second = 117.546
eval_steps_per_second = 0.92
eval_total = 11548
***** Eval results *****
2021-12-14 15:41:40.429066epoch = 7.0
eval_HasAns_exact = 83.47765846899897
eval_HasAns_f1 = 88.97497847186122
eval_HasAns_total = 5774
eval_NoAns_exact = 83.28714928992034
eval_NoAns_f1 = 83.28714928992034
eval_NoAns_total = 5774
eval_best_exact = 87.56494631104954
eval_best_exact_thresh = -5.70703125
eval_best_f1 = 90.1712271871601
eval_best_f1_thresh = -5.6875
eval_exact = 83.38240387945964
eval_f1 = 86.13106388089054
eval_runtime = 116.1628
eval_samples_per_second = 119.866
eval_steps_per_second = 0.938
eval_total = 11548
***** Eval results *****
2021-12-14 18:00:18.739183epoch = 8.0
eval_HasAns_exact = 83.80671977831659
eval_HasAns_f1 = 89.25279748539205
eval_HasAns_total = 5774
eval_NoAns_exact = 82.35192241080706
eval_NoAns_f1 = 82.35192241080706
eval_NoAns_total = 5774
eval_best_exact = 87.7641149982681
eval_best_exact_thresh = -5.828125
eval_best_f1 = 90.35660938760992
eval_best_f1_thresh = -5.828125
eval_exact = 83.07932109456183
eval_f1 = 85.8023599480993
eval_runtime = 114.8772
eval_samples_per_second = 121.208
eval_steps_per_second = 0.949
eval_total = 11548
***** Eval results *****
2021-12-14 20:19:11.241603epoch = 9.0
eval_HasAns_exact = 83.70280568063734
eval_HasAns_f1 = 89.21377471775827
eval_HasAns_total = 5774
eval_NoAns_exact = 82.38656044336682
eval_NoAns_f1 = 82.38656044336682
eval_NoAns_total = 5774
eval_best_exact = 87.85936958780741
eval_best_exact_thresh = -6.28515625
eval_best_f1 = 90.4998967568238
eval_best_f1_thresh = -5.98046875
eval_exact = 83.04468306200208
eval_f1 = 85.80016758056232
eval_runtime = 114.8108
eval_samples_per_second = 121.278
eval_steps_per_second = 0.949
eval_total = 11548
***** Eval results *****
2021-12-14 22:37:47.534128epoch = 10.0
eval_HasAns_exact = 83.65084863179771
eval_HasAns_f1 = 89.1463632488054
eval_HasAns_total = 5774
eval_NoAns_exact = 82.99272601316245
eval_NoAns_f1 = 82.99272601316245
eval_NoAns_total = 5774
eval_best_exact = 87.81607204710772
eval_best_exact_thresh = -6.37890625
eval_best_f1 = 90.44887200606756
eval_best_f1_thresh = -6.046875
eval_exact = 83.32178732248008
eval_f1 = 86.0695446309837
eval_runtime = 116.3261
eval_samples_per_second = 119.698
eval_steps_per_second = 0.937
eval_total = 11548