<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
bert-base-uncased_winobias_classifieronly
This model is a fine-tuned version of bert-base-uncased on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.6924
- Accuracy: 0.5290
- Tp: 0.2614
- Tn: 0.2677
- Fp: 0.2323
- Fn: 0.2386
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 0.0001
- train_batch_size: 64
- eval_batch_size: 64
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 50
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy | Tp | Tn | Fp | Fn |
---|---|---|---|---|---|---|---|---|
0.6962 | 0.8 | 20 | 0.6948 | 0.5019 | 0.2203 | 0.2816 | 0.2184 | 0.2797 |
0.6962 | 1.6 | 40 | 0.6944 | 0.4981 | 0.2696 | 0.2285 | 0.2715 | 0.2304 |
0.6967 | 2.4 | 60 | 0.6954 | 0.5019 | 0.4078 | 0.0941 | 0.4059 | 0.0922 |
0.6963 | 3.2 | 80 | 0.6944 | 0.5088 | 0.0581 | 0.4508 | 0.0492 | 0.4419 |
0.6949 | 4.0 | 100 | 0.6936 | 0.4987 | 0.2948 | 0.2039 | 0.2961 | 0.2052 |
0.7003 | 4.8 | 120 | 0.6944 | 0.5013 | 0.0322 | 0.4691 | 0.0309 | 0.4678 |
0.6948 | 5.6 | 140 | 0.6938 | 0.4987 | 0.3630 | 0.1357 | 0.3643 | 0.1370 |
0.695 | 6.4 | 160 | 0.6945 | 0.5032 | 0.0284 | 0.4747 | 0.0253 | 0.4716 |
0.6954 | 7.2 | 180 | 0.6939 | 0.5025 | 0.0739 | 0.4287 | 0.0713 | 0.4261 |
0.6914 | 8.0 | 200 | 0.6936 | 0.5019 | 0.3491 | 0.1528 | 0.3472 | 0.1509 |
0.6968 | 8.8 | 220 | 0.6945 | 0.4981 | 0.0234 | 0.4747 | 0.0253 | 0.4766 |
0.6961 | 9.6 | 240 | 0.6934 | 0.5069 | 0.2020 | 0.3049 | 0.1951 | 0.2980 |
0.6957 | 10.4 | 260 | 0.6932 | 0.5038 | 0.2917 | 0.2121 | 0.2879 | 0.2083 |
0.6929 | 11.2 | 280 | 0.6939 | 0.5032 | 0.0271 | 0.4760 | 0.0240 | 0.4729 |
0.7065 | 12.0 | 300 | 0.6934 | 0.4987 | 0.4009 | 0.0979 | 0.4021 | 0.0991 |
0.6986 | 12.8 | 320 | 0.6934 | 0.5025 | 0.0833 | 0.4192 | 0.0808 | 0.4167 |
0.6974 | 13.6 | 340 | 0.6945 | 0.4949 | 0.0221 | 0.4729 | 0.0271 | 0.4779 |
0.6991 | 14.4 | 360 | 0.6944 | 0.5044 | 0.4956 | 0.0088 | 0.4912 | 0.0044 |
0.6961 | 15.2 | 380 | 0.6932 | 0.5038 | 0.0751 | 0.4287 | 0.0713 | 0.4249 |
0.6982 | 16.0 | 400 | 0.6928 | 0.5057 | 0.1641 | 0.3415 | 0.1585 | 0.3359 |
0.6986 | 16.8 | 420 | 0.6928 | 0.5101 | 0.1351 | 0.375 | 0.125 | 0.3649 |
0.695 | 17.6 | 440 | 0.6927 | 0.5107 | 0.2702 | 0.2405 | 0.2595 | 0.2298 |
0.6954 | 18.4 | 460 | 0.6928 | 0.5095 | 0.3441 | 0.1654 | 0.3346 | 0.1559 |
0.6954 | 19.2 | 480 | 0.6926 | 0.5133 | 0.2626 | 0.2506 | 0.2494 | 0.2374 |
0.7003 | 20.0 | 500 | 0.6926 | 0.5183 | 0.2708 | 0.2475 | 0.2525 | 0.2292 |
0.6982 | 20.8 | 520 | 0.6926 | 0.5082 | 0.2652 | 0.2431 | 0.2569 | 0.2348 |
0.696 | 21.6 | 540 | 0.6927 | 0.5170 | 0.2456 | 0.2715 | 0.2285 | 0.2544 |
0.6928 | 22.4 | 560 | 0.6928 | 0.5038 | 0.3958 | 0.1080 | 0.3920 | 0.1042 |
0.6968 | 23.2 | 580 | 0.6927 | 0.5164 | 0.3062 | 0.2102 | 0.2898 | 0.1938 |
0.693 | 24.0 | 600 | 0.6926 | 0.5196 | 0.2702 | 0.2494 | 0.2506 | 0.2298 |
0.7 | 24.8 | 620 | 0.6927 | 0.5088 | 0.4148 | 0.0941 | 0.4059 | 0.0852 |
0.6993 | 25.6 | 640 | 0.6932 | 0.5044 | 0.0335 | 0.4710 | 0.0290 | 0.4665 |
0.7022 | 26.4 | 660 | 0.6926 | 0.5152 | 0.1806 | 0.3346 | 0.1654 | 0.3194 |
0.6954 | 27.2 | 680 | 0.6925 | 0.5170 | 0.2241 | 0.2929 | 0.2071 | 0.2759 |
0.6978 | 28.0 | 700 | 0.6925 | 0.5158 | 0.3289 | 0.1869 | 0.3131 | 0.1711 |
0.7013 | 28.8 | 720 | 0.6925 | 0.5145 | 0.2601 | 0.2544 | 0.2456 | 0.2399 |
0.7004 | 29.6 | 740 | 0.6931 | 0.5063 | 0.4729 | 0.0335 | 0.4665 | 0.0271 |
0.6913 | 30.4 | 760 | 0.6926 | 0.5164 | 0.1982 | 0.3182 | 0.1818 | 0.3018 |
0.6974 | 31.2 | 780 | 0.6926 | 0.5164 | 0.1900 | 0.3264 | 0.1736 | 0.3100 |
0.6904 | 32.0 | 800 | 0.6926 | 0.5107 | 0.3933 | 0.1174 | 0.3826 | 0.1067 |
0.6917 | 32.8 | 820 | 0.6925 | 0.5271 | 0.2961 | 0.2311 | 0.2689 | 0.2039 |
0.6956 | 33.6 | 840 | 0.6927 | 0.5114 | 0.1098 | 0.4015 | 0.0985 | 0.3902 |
0.6941 | 34.4 | 860 | 0.6927 | 0.5133 | 0.4350 | 0.0783 | 0.4217 | 0.0650 |
0.6935 | 35.2 | 880 | 0.6928 | 0.5114 | 0.4571 | 0.0543 | 0.4457 | 0.0429 |
0.6935 | 36.0 | 900 | 0.6925 | 0.5114 | 0.3643 | 0.1471 | 0.3529 | 0.1357 |
0.6935 | 36.8 | 920 | 0.6924 | 0.5177 | 0.2020 | 0.3157 | 0.1843 | 0.2980 |
0.6916 | 37.6 | 940 | 0.6925 | 0.5177 | 0.3314 | 0.1862 | 0.3138 | 0.1686 |
0.703 | 38.4 | 960 | 0.6925 | 0.5183 | 0.3220 | 0.1963 | 0.3037 | 0.1780 |
0.698 | 39.2 | 980 | 0.6925 | 0.5234 | 0.3005 | 0.2229 | 0.2771 | 0.1995 |
0.6966 | 40.0 | 1000 | 0.6925 | 0.5240 | 0.3150 | 0.2090 | 0.2910 | 0.1850 |
0.6942 | 40.8 | 1020 | 0.6925 | 0.5290 | 0.3049 | 0.2241 | 0.2759 | 0.1951 |
0.6916 | 41.6 | 1040 | 0.6924 | 0.5240 | 0.2999 | 0.2241 | 0.2759 | 0.2001 |
0.6946 | 42.4 | 1060 | 0.6925 | 0.5221 | 0.1862 | 0.3359 | 0.1641 | 0.3138 |
0.6925 | 43.2 | 1080 | 0.6926 | 0.5025 | 0.1155 | 0.3870 | 0.1130 | 0.3845 |
0.6977 | 44.0 | 1100 | 0.6925 | 0.5170 | 0.1559 | 0.3611 | 0.1389 | 0.3441 |
0.6938 | 44.8 | 1120 | 0.6924 | 0.5221 | 0.2008 | 0.3213 | 0.1787 | 0.2992 |
0.6972 | 45.6 | 1140 | 0.6924 | 0.5208 | 0.2449 | 0.2759 | 0.2241 | 0.2551 |
0.6938 | 46.4 | 1160 | 0.6924 | 0.5227 | 0.3213 | 0.2014 | 0.2986 | 0.1787 |
0.6924 | 47.2 | 1180 | 0.6924 | 0.5107 | 0.3396 | 0.1711 | 0.3289 | 0.1604 |
0.7034 | 48.0 | 1200 | 0.6924 | 0.5322 | 0.2980 | 0.2342 | 0.2658 | 0.2020 |
0.6935 | 48.8 | 1220 | 0.6924 | 0.5316 | 0.2809 | 0.2506 | 0.2494 | 0.2191 |
0.6971 | 49.6 | 1240 | 0.6924 | 0.5290 | 0.2614 | 0.2677 | 0.2323 | 0.2386 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.13.1
- Datasets 2.10.1
- Tokenizers 0.13.2