|
--- |
|
license: apache-2.0 |
|
tags: |
|
- generated_from_trainer |
|
datasets: |
|
- stereoset |
|
metrics: |
|
- accuracy |
|
model-index: |
|
- name: multiberts-seed_1_stereoset_classifieronly |
|
results: |
|
- task: |
|
name: Text Classification |
|
type: text-classification |
|
dataset: |
|
name: stereoset |
|
type: stereoset |
|
config: intersentence |
|
split: validation |
|
args: intersentence |
|
metrics: |
|
- name: Accuracy |
|
type: accuracy |
|
value: 0.5596546310832025 |
|
--- |
|
|
|
<!-- 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. --> |
|
|
|
# multiberts-seed_1_stereoset_classifieronly |
|
|
|
This model is a fine-tuned version of [google/multiberts-seed_1](https://huggingface.co/google/multiberts-seed_1) on the stereoset dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.6867 |
|
- Accuracy: 0.5597 |
|
- Tp: 0.3234 |
|
- Tn: 0.2363 |
|
- Fp: 0.2763 |
|
- Fn: 0.1641 |
|
|
|
## 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: 5e-05 |
|
- 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 | |
|
|:-------------:|:-----:|:----:|:---------------:|:--------:|:------:|:------:|:------:|:------:| |
|
| 0.7062 | 0.43 | 20 | 0.6925 | 0.5220 | 0.2951 | 0.2268 | 0.2857 | 0.1923 | |
|
| 0.6988 | 0.85 | 40 | 0.6985 | 0.5181 | 0.4082 | 0.1099 | 0.4027 | 0.0793 | |
|
| 0.7023 | 1.28 | 60 | 0.6920 | 0.5173 | 0.2951 | 0.2221 | 0.2904 | 0.1923 | |
|
| 0.6934 | 1.7 | 80 | 0.6944 | 0.5196 | 0.3422 | 0.1774 | 0.3352 | 0.1452 | |
|
| 0.6983 | 2.13 | 100 | 0.6918 | 0.5204 | 0.3093 | 0.2111 | 0.3014 | 0.1782 | |
|
| 0.7046 | 2.55 | 120 | 0.6922 | 0.5188 | 0.3265 | 0.1923 | 0.3203 | 0.1609 | |
|
| 0.7051 | 2.98 | 140 | 0.6912 | 0.5283 | 0.2111 | 0.3171 | 0.1954 | 0.2763 | |
|
| 0.6994 | 3.4 | 160 | 0.6944 | 0.5196 | 0.3540 | 0.1656 | 0.3469 | 0.1334 | |
|
| 0.6914 | 3.83 | 180 | 0.6962 | 0.5212 | 0.3854 | 0.1358 | 0.3768 | 0.1020 | |
|
| 0.7066 | 4.26 | 200 | 0.6907 | 0.5330 | 0.2347 | 0.2983 | 0.2143 | 0.2527 | |
|
| 0.6975 | 4.68 | 220 | 0.6914 | 0.5345 | 0.2904 | 0.2441 | 0.2684 | 0.1970 | |
|
| 0.6995 | 5.11 | 240 | 0.6923 | 0.5275 | 0.3289 | 0.1986 | 0.3140 | 0.1586 | |
|
| 0.7011 | 5.53 | 260 | 0.6909 | 0.5377 | 0.2881 | 0.2496 | 0.2630 | 0.1994 | |
|
| 0.6985 | 5.96 | 280 | 0.6924 | 0.5251 | 0.3407 | 0.1845 | 0.3281 | 0.1468 | |
|
| 0.6957 | 6.38 | 300 | 0.6915 | 0.5275 | 0.3281 | 0.1994 | 0.3132 | 0.1593 | |
|
| 0.7033 | 6.81 | 320 | 0.6904 | 0.5400 | 0.2889 | 0.2512 | 0.2614 | 0.1986 | |
|
| 0.6989 | 7.23 | 340 | 0.6900 | 0.5392 | 0.2394 | 0.2998 | 0.2127 | 0.2480 | |
|
| 0.6913 | 7.66 | 360 | 0.6906 | 0.5353 | 0.3203 | 0.2151 | 0.2975 | 0.1672 | |
|
| 0.6906 | 8.09 | 380 | 0.6950 | 0.5290 | 0.3721 | 0.1570 | 0.3556 | 0.1154 | |
|
| 0.6854 | 8.51 | 400 | 0.6908 | 0.5338 | 0.2873 | 0.2465 | 0.2661 | 0.2002 | |
|
| 0.6888 | 8.94 | 420 | 0.6948 | 0.5290 | 0.3689 | 0.1601 | 0.3524 | 0.1185 | |
|
| 0.7005 | 9.36 | 440 | 0.6935 | 0.5330 | 0.3485 | 0.1845 | 0.3281 | 0.1389 | |
|
| 0.695 | 9.79 | 460 | 0.6899 | 0.5392 | 0.2465 | 0.2928 | 0.2198 | 0.2410 | |
|
| 0.69 | 10.21 | 480 | 0.6916 | 0.5259 | 0.2896 | 0.2363 | 0.2763 | 0.1978 | |
|
| 0.6904 | 10.64 | 500 | 0.6918 | 0.5275 | 0.3022 | 0.2253 | 0.2873 | 0.1852 | |
|
| 0.6802 | 11.06 | 520 | 0.6928 | 0.5235 | 0.3155 | 0.2080 | 0.3046 | 0.1719 | |
|
| 0.6986 | 11.49 | 540 | 0.6918 | 0.5275 | 0.3077 | 0.2198 | 0.2928 | 0.1797 | |
|
| 0.697 | 11.91 | 560 | 0.6895 | 0.5408 | 0.2637 | 0.2771 | 0.2355 | 0.2237 | |
|
| 0.6957 | 12.34 | 580 | 0.6916 | 0.5361 | 0.3234 | 0.2127 | 0.2998 | 0.1641 | |
|
| 0.6967 | 12.77 | 600 | 0.6932 | 0.5267 | 0.3407 | 0.1860 | 0.3265 | 0.1468 | |
|
| 0.6915 | 13.19 | 620 | 0.6907 | 0.5361 | 0.2818 | 0.2543 | 0.2582 | 0.2057 | |
|
| 0.686 | 13.62 | 640 | 0.6925 | 0.5306 | 0.3328 | 0.1978 | 0.3148 | 0.1546 | |
|
| 0.6972 | 14.04 | 660 | 0.6900 | 0.5385 | 0.2818 | 0.2567 | 0.2559 | 0.2057 | |
|
| 0.687 | 14.47 | 680 | 0.6891 | 0.5385 | 0.2441 | 0.2943 | 0.2182 | 0.2433 | |
|
| 0.7001 | 14.89 | 700 | 0.6885 | 0.5338 | 0.2378 | 0.2959 | 0.2166 | 0.2496 | |
|
| 0.6925 | 15.32 | 720 | 0.6887 | 0.5408 | 0.2951 | 0.2457 | 0.2669 | 0.1923 | |
|
| 0.6956 | 15.74 | 740 | 0.6904 | 0.5408 | 0.3414 | 0.1994 | 0.3132 | 0.1460 | |
|
| 0.6825 | 16.17 | 760 | 0.6892 | 0.5345 | 0.2967 | 0.2378 | 0.2747 | 0.1907 | |
|
| 0.6872 | 16.6 | 780 | 0.6889 | 0.5416 | 0.3022 | 0.2394 | 0.2732 | 0.1852 | |
|
| 0.6916 | 17.02 | 800 | 0.6913 | 0.5416 | 0.3462 | 0.1954 | 0.3171 | 0.1413 | |
|
| 0.6891 | 17.45 | 820 | 0.6909 | 0.5361 | 0.3336 | 0.2025 | 0.3100 | 0.1538 | |
|
| 0.696 | 17.87 | 840 | 0.6881 | 0.5463 | 0.2630 | 0.2834 | 0.2292 | 0.2245 | |
|
| 0.6926 | 18.3 | 860 | 0.6886 | 0.5432 | 0.2865 | 0.2567 | 0.2559 | 0.2009 | |
|
| 0.6867 | 18.72 | 880 | 0.6889 | 0.5416 | 0.2998 | 0.2418 | 0.2708 | 0.1876 | |
|
| 0.6957 | 19.15 | 900 | 0.6883 | 0.5447 | 0.2889 | 0.2559 | 0.2567 | 0.1986 | |
|
| 0.6753 | 19.57 | 920 | 0.6884 | 0.5432 | 0.2983 | 0.2449 | 0.2677 | 0.1892 | |
|
| 0.7012 | 20.0 | 940 | 0.6903 | 0.5416 | 0.3438 | 0.1978 | 0.3148 | 0.1436 | |
|
| 0.6994 | 20.43 | 960 | 0.6883 | 0.5432 | 0.2959 | 0.2473 | 0.2653 | 0.1915 | |
|
| 0.6871 | 20.85 | 980 | 0.6889 | 0.5447 | 0.3265 | 0.2182 | 0.2943 | 0.1609 | |
|
| 0.6946 | 21.28 | 1000 | 0.6892 | 0.5479 | 0.3297 | 0.2182 | 0.2943 | 0.1578 | |
|
| 0.6853 | 21.7 | 1020 | 0.6887 | 0.5455 | 0.3226 | 0.2229 | 0.2896 | 0.1648 | |
|
| 0.6988 | 22.13 | 1040 | 0.6900 | 0.5455 | 0.3509 | 0.1947 | 0.3179 | 0.1366 | |
|
| 0.6933 | 22.55 | 1060 | 0.6873 | 0.5455 | 0.2661 | 0.2794 | 0.2331 | 0.2214 | |
|
| 0.6868 | 22.98 | 1080 | 0.6899 | 0.5432 | 0.3556 | 0.1876 | 0.3250 | 0.1319 | |
|
| 0.6931 | 23.4 | 1100 | 0.6905 | 0.5400 | 0.3619 | 0.1782 | 0.3344 | 0.1256 | |
|
| 0.6907 | 23.83 | 1120 | 0.6869 | 0.5502 | 0.2912 | 0.2590 | 0.2535 | 0.1962 | |
|
| 0.6848 | 24.26 | 1140 | 0.6878 | 0.5502 | 0.3163 | 0.2339 | 0.2786 | 0.1711 | |
|
| 0.7056 | 24.68 | 1160 | 0.6887 | 0.5487 | 0.3336 | 0.2151 | 0.2975 | 0.1538 | |
|
| 0.6797 | 25.11 | 1180 | 0.6878 | 0.5447 | 0.3077 | 0.2370 | 0.2755 | 0.1797 | |
|
| 0.6891 | 25.53 | 1200 | 0.6885 | 0.5463 | 0.3179 | 0.2284 | 0.2841 | 0.1695 | |
|
| 0.6945 | 25.96 | 1220 | 0.6892 | 0.5495 | 0.3399 | 0.2096 | 0.3030 | 0.1476 | |
|
| 0.6798 | 26.38 | 1240 | 0.6907 | 0.5440 | 0.3603 | 0.1837 | 0.3289 | 0.1272 | |
|
| 0.7025 | 26.81 | 1260 | 0.6872 | 0.5495 | 0.2943 | 0.2551 | 0.2575 | 0.1931 | |
|
| 0.6903 | 27.23 | 1280 | 0.6870 | 0.5424 | 0.2669 | 0.2755 | 0.2370 | 0.2206 | |
|
| 0.6995 | 27.66 | 1300 | 0.6885 | 0.5487 | 0.3359 | 0.2127 | 0.2998 | 0.1515 | |
|
| 0.6916 | 28.09 | 1320 | 0.6872 | 0.5495 | 0.3046 | 0.2449 | 0.2677 | 0.1829 | |
|
| 0.695 | 28.51 | 1340 | 0.6866 | 0.5463 | 0.2559 | 0.2904 | 0.2221 | 0.2316 | |
|
| 0.6864 | 28.94 | 1360 | 0.6875 | 0.5495 | 0.3226 | 0.2268 | 0.2857 | 0.1648 | |
|
| 0.6932 | 29.36 | 1380 | 0.6870 | 0.5510 | 0.3100 | 0.2410 | 0.2716 | 0.1774 | |
|
| 0.6876 | 29.79 | 1400 | 0.6866 | 0.5510 | 0.3022 | 0.2488 | 0.2637 | 0.1852 | |
|
| 0.6923 | 30.21 | 1420 | 0.6878 | 0.5495 | 0.3312 | 0.2182 | 0.2943 | 0.1562 | |
|
| 0.6954 | 30.64 | 1440 | 0.6906 | 0.5432 | 0.3681 | 0.1750 | 0.3375 | 0.1193 | |
|
| 0.699 | 31.06 | 1460 | 0.6891 | 0.5502 | 0.3516 | 0.1986 | 0.3140 | 0.1358 | |
|
| 0.6864 | 31.49 | 1480 | 0.6877 | 0.5542 | 0.3140 | 0.2402 | 0.2724 | 0.1735 | |
|
| 0.6952 | 31.91 | 1500 | 0.6868 | 0.5518 | 0.2818 | 0.2700 | 0.2425 | 0.2057 | |
|
| 0.6936 | 32.34 | 1520 | 0.6870 | 0.5510 | 0.3022 | 0.2488 | 0.2637 | 0.1852 | |
|
| 0.6892 | 32.77 | 1540 | 0.6891 | 0.5463 | 0.3524 | 0.1939 | 0.3187 | 0.1350 | |
|
| 0.6891 | 33.19 | 1560 | 0.6884 | 0.5463 | 0.3430 | 0.2033 | 0.3093 | 0.1444 | |
|
| 0.6836 | 33.62 | 1580 | 0.6874 | 0.5534 | 0.3375 | 0.2159 | 0.2967 | 0.1499 | |
|
| 0.6824 | 34.04 | 1600 | 0.6869 | 0.5526 | 0.3140 | 0.2386 | 0.2739 | 0.1735 | |
|
| 0.6834 | 34.47 | 1620 | 0.6894 | 0.5440 | 0.3603 | 0.1837 | 0.3289 | 0.1272 | |
|
| 0.6961 | 34.89 | 1640 | 0.6874 | 0.5526 | 0.3265 | 0.2261 | 0.2865 | 0.1609 | |
|
| 0.6935 | 35.32 | 1660 | 0.6862 | 0.5479 | 0.2684 | 0.2794 | 0.2331 | 0.2190 | |
|
| 0.6881 | 35.74 | 1680 | 0.6867 | 0.5479 | 0.2967 | 0.2512 | 0.2614 | 0.1907 | |
|
| 0.6935 | 36.17 | 1700 | 0.6883 | 0.5502 | 0.3383 | 0.2119 | 0.3006 | 0.1491 | |
|
| 0.6874 | 36.6 | 1720 | 0.6873 | 0.5565 | 0.3179 | 0.2386 | 0.2739 | 0.1695 | |
|
| 0.683 | 37.02 | 1740 | 0.6872 | 0.5549 | 0.3163 | 0.2386 | 0.2739 | 0.1711 | |
|
| 0.6931 | 37.45 | 1760 | 0.6871 | 0.5581 | 0.3234 | 0.2347 | 0.2779 | 0.1641 | |
|
| 0.6859 | 37.87 | 1780 | 0.6877 | 0.5479 | 0.3367 | 0.2111 | 0.3014 | 0.1507 | |
|
| 0.6877 | 38.3 | 1800 | 0.6882 | 0.5455 | 0.3438 | 0.2017 | 0.3108 | 0.1436 | |
|
| 0.6938 | 38.72 | 1820 | 0.6883 | 0.5471 | 0.3462 | 0.2009 | 0.3116 | 0.1413 | |
|
| 0.694 | 39.15 | 1840 | 0.6868 | 0.5565 | 0.3148 | 0.2418 | 0.2708 | 0.1727 | |
|
| 0.6938 | 39.57 | 1860 | 0.6862 | 0.5542 | 0.2975 | 0.2567 | 0.2559 | 0.1900 | |
|
| 0.6883 | 40.0 | 1880 | 0.6865 | 0.5526 | 0.3030 | 0.2496 | 0.2630 | 0.1845 | |
|
| 0.6867 | 40.43 | 1900 | 0.6868 | 0.5526 | 0.3124 | 0.2402 | 0.2724 | 0.1750 | |
|
| 0.6933 | 40.85 | 1920 | 0.6867 | 0.5542 | 0.3132 | 0.2410 | 0.2716 | 0.1743 | |
|
| 0.6915 | 41.28 | 1940 | 0.6868 | 0.5557 | 0.3163 | 0.2394 | 0.2732 | 0.1711 | |
|
| 0.6905 | 41.7 | 1960 | 0.6865 | 0.5542 | 0.3116 | 0.2425 | 0.2700 | 0.1758 | |
|
| 0.7036 | 42.13 | 1980 | 0.6864 | 0.5510 | 0.3069 | 0.2441 | 0.2684 | 0.1805 | |
|
| 0.6892 | 42.55 | 2000 | 0.6866 | 0.5534 | 0.3108 | 0.2425 | 0.2700 | 0.1766 | |
|
| 0.6811 | 42.98 | 2020 | 0.6868 | 0.5581 | 0.3226 | 0.2355 | 0.2771 | 0.1648 | |
|
| 0.6925 | 43.4 | 2040 | 0.6875 | 0.5526 | 0.3383 | 0.2143 | 0.2983 | 0.1491 | |
|
| 0.6963 | 43.83 | 2060 | 0.6877 | 0.5502 | 0.3422 | 0.2080 | 0.3046 | 0.1452 | |
|
| 0.6887 | 44.26 | 2080 | 0.6869 | 0.5597 | 0.3250 | 0.2347 | 0.2779 | 0.1625 | |
|
| 0.6896 | 44.68 | 2100 | 0.6870 | 0.5589 | 0.3250 | 0.2339 | 0.2786 | 0.1625 | |
|
| 0.6826 | 45.11 | 2120 | 0.6872 | 0.5597 | 0.3312 | 0.2284 | 0.2841 | 0.1562 | |
|
| 0.6823 | 45.53 | 2140 | 0.6873 | 0.5565 | 0.3328 | 0.2237 | 0.2889 | 0.1546 | |
|
| 0.6982 | 45.96 | 2160 | 0.6872 | 0.5557 | 0.3305 | 0.2253 | 0.2873 | 0.1570 | |
|
| 0.6874 | 46.38 | 2180 | 0.6870 | 0.5597 | 0.3281 | 0.2316 | 0.2810 | 0.1593 | |
|
| 0.6887 | 46.81 | 2200 | 0.6867 | 0.5573 | 0.3226 | 0.2347 | 0.2779 | 0.1648 | |
|
| 0.6967 | 47.23 | 2220 | 0.6868 | 0.5589 | 0.3250 | 0.2339 | 0.2786 | 0.1625 | |
|
| 0.6867 | 47.66 | 2240 | 0.6869 | 0.5581 | 0.3265 | 0.2316 | 0.2810 | 0.1609 | |
|
| 0.678 | 48.09 | 2260 | 0.6868 | 0.5581 | 0.3257 | 0.2323 | 0.2802 | 0.1617 | |
|
| 0.6847 | 48.51 | 2280 | 0.6867 | 0.5597 | 0.3250 | 0.2347 | 0.2779 | 0.1625 | |
|
| 0.6933 | 48.94 | 2300 | 0.6867 | 0.5604 | 0.3242 | 0.2363 | 0.2763 | 0.1633 | |
|
| 0.6847 | 49.36 | 2320 | 0.6866 | 0.5612 | 0.3234 | 0.2378 | 0.2747 | 0.1641 | |
|
| 0.6906 | 49.79 | 2340 | 0.6867 | 0.5597 | 0.3234 | 0.2363 | 0.2763 | 0.1641 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.26.1 |
|
- Pytorch 1.13.1 |
|
- Datasets 2.10.1 |
|
- Tokenizers 0.13.2 |
|
|
