<!-- 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. -->
ellis-v1-emotion-regency
This model is a fine-tuned version of distilbert-base-uncased on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.1698
- Accuracy: 0.9807
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: 2e-05
- train_batch_size: 16
- eval_batch_size: 16
- 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 |
---|---|---|---|---|
No log | 1.0 | 495 | 0.3350 | 0.9455 |
1.5244 | 2.0 | 990 | 0.1552 | 0.9636 |
0.2196 | 3.0 | 1485 | 0.1216 | 0.9727 |
0.0776 | 4.0 | 1980 | 0.0958 | 0.975 |
0.0397 | 5.0 | 2475 | 0.1276 | 0.9716 |
0.0288 | 6.0 | 2970 | 0.1385 | 0.9739 |
0.0179 | 7.0 | 3465 | 0.1580 | 0.975 |
0.0093 | 8.0 | 3960 | 0.1463 | 0.9727 |
0.0131 | 9.0 | 4455 | 0.1235 | 0.975 |
0.0107 | 10.0 | 4950 | 0.1803 | 0.9773 |
0.0056 | 11.0 | 5445 | 0.1679 | 0.9784 |
0.0079 | 12.0 | 5940 | 0.1834 | 0.9739 |
0.0034 | 13.0 | 6435 | 0.1740 | 0.9739 |
0.0099 | 14.0 | 6930 | 0.1611 | 0.9682 |
0.0059 | 15.0 | 7425 | 0.1638 | 0.9761 |
0.0145 | 16.0 | 7920 | 0.1733 | 0.9761 |
0.0044 | 17.0 | 8415 | 0.1478 | 0.9795 |
0.0069 | 18.0 | 8910 | 0.1848 | 0.9773 |
0.0057 | 19.0 | 9405 | 0.1810 | 0.9727 |
0.0053 | 20.0 | 9900 | 0.1595 | 0.9773 |
0.0105 | 21.0 | 10395 | 0.1698 | 0.9761 |
0.0025 | 22.0 | 10890 | 0.1575 | 0.9761 |
0.0032 | 23.0 | 11385 | 0.1736 | 0.9761 |
0.0046 | 24.0 | 11880 | 0.1637 | 0.9761 |
0.0002 | 25.0 | 12375 | 0.2291 | 0.9739 |
0.002 | 26.0 | 12870 | 0.1375 | 0.9830 |
0.0001 | 27.0 | 13365 | 0.2430 | 0.9739 |
0.0031 | 28.0 | 13860 | 0.1825 | 0.9773 |
0.0003 | 29.0 | 14355 | 0.1787 | 0.9773 |
0.0013 | 30.0 | 14850 | 0.2311 | 0.9739 |
0.0022 | 31.0 | 15345 | 0.2732 | 0.9693 |
0.0038 | 32.0 | 15840 | 0.1949 | 0.9784 |
0.0017 | 33.0 | 16335 | 0.1866 | 0.9795 |
0.0009 | 34.0 | 16830 | 0.2008 | 0.9784 |
0.0013 | 35.0 | 17325 | 0.1873 | 0.9807 |
0.0009 | 36.0 | 17820 | 0.1615 | 0.9841 |
0.0026 | 37.0 | 18315 | 0.1879 | 0.9773 |
0.0005 | 38.0 | 18810 | 0.1986 | 0.9784 |
0.0 | 39.0 | 19305 | 0.1891 | 0.9795 |
0.0002 | 40.0 | 19800 | 0.1781 | 0.9818 |
0.0 | 41.0 | 20295 | 0.1775 | 0.9807 |
0.0 | 42.0 | 20790 | 0.1711 | 0.9807 |
0.0 | 43.0 | 21285 | 0.1745 | 0.9830 |
0.0015 | 44.0 | 21780 | 0.1750 | 0.9830 |
0.0 | 45.0 | 22275 | 0.1720 | 0.9784 |
0.0 | 46.0 | 22770 | 0.1726 | 0.9784 |
0.0 | 47.0 | 23265 | 0.1734 | 0.9773 |
0.0 | 48.0 | 23760 | 0.1738 | 0.9773 |
0.0001 | 49.0 | 24255 | 0.1698 | 0.9807 |
0.0 | 50.0 | 24750 | 0.1698 | 0.9807 |
Framework versions
- Transformers 4.31.0
- Pytorch 2.0.1+cu118
- Datasets 2.13.0
- Tokenizers 0.13.3