<!-- 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. -->
TUF_DistilBERT_5E
This model is a fine-tuned version of distilbert-base-cased on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.1832
- Accuracy: 0.96
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: 1e-05
- train_batch_size: 16
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 0.5092 | 0.1 | 50 | 0.4385 | 0.7533 |
| 0.2807 | 0.2 | 100 | 0.2225 | 0.9 |
| 0.1881 | 0.3 | 150 | 0.1531 | 0.94 |
| 0.1895 | 0.4 | 200 | 0.1426 | 0.94 |
| 0.1995 | 0.5 | 250 | 0.1428 | 0.94 |
| 0.1745 | 0.59 | 300 | 0.1538 | 0.9267 |
| 0.1679 | 0.69 | 350 | 0.1249 | 0.9533 |
| 0.199 | 0.79 | 400 | 0.1327 | 0.9467 |
| 0.1703 | 0.89 | 450 | 0.1488 | 0.92 |
| 0.1541 | 0.99 | 500 | 0.1772 | 0.9467 |
| 0.1436 | 1.09 | 550 | 0.1070 | 0.9667 |
| 0.1463 | 1.19 | 600 | 0.1165 | 0.9467 |
| 0.1309 | 1.29 | 650 | 0.1054 | 0.9733 |
| 0.097 | 1.39 | 700 | 0.1346 | 0.94 |
| 0.1307 | 1.49 | 750 | 0.1477 | 0.9467 |
| 0.1506 | 1.58 | 800 | 0.1311 | 0.9533 |
| 0.1386 | 1.68 | 850 | 0.1165 | 0.9667 |
| 0.1463 | 1.78 | 900 | 0.4207 | 0.9067 |
| 0.1202 | 1.88 | 950 | 0.1528 | 0.9667 |
| 0.1403 | 1.98 | 1000 | 0.1262 | 0.96 |
| 0.073 | 2.08 | 1050 | 0.1459 | 0.96 |
| 0.0713 | 2.18 | 1100 | 0.1747 | 0.9533 |
| 0.0814 | 2.28 | 1150 | 0.1953 | 0.9667 |
| 0.0935 | 2.38 | 1200 | 0.1888 | 0.9533 |
| 0.0685 | 2.48 | 1250 | 0.1562 | 0.9467 |
| 0.1154 | 2.57 | 1300 | 0.1806 | 0.96 |
| 0.1239 | 2.67 | 1350 | 0.1322 | 0.9533 |
| 0.1011 | 2.77 | 1400 | 0.2148 | 0.94 |
| 0.0718 | 2.87 | 1450 | 0.1686 | 0.96 |
| 0.1159 | 2.97 | 1500 | 0.1532 | 0.9533 |
| 0.0516 | 3.07 | 1550 | 0.1888 | 0.96 |
| 0.063 | 3.17 | 1600 | 0.1851 | 0.9467 |
| 0.068 | 3.27 | 1650 | 0.2775 | 0.94 |
| 0.0946 | 3.37 | 1700 | 0.1853 | 0.96 |
| 0.0606 | 3.47 | 1750 | 0.2148 | 0.9467 |
| 0.0663 | 3.56 | 1800 | 0.2091 | 0.9533 |
| 0.0474 | 3.66 | 1850 | 0.1702 | 0.9533 |
| 0.0585 | 3.76 | 1900 | 0.1660 | 0.96 |
| 0.0439 | 3.86 | 1950 | 0.2220 | 0.9533 |
| 0.0758 | 3.96 | 2000 | 0.1834 | 0.96 |
| 0.0497 | 4.06 | 2050 | 0.1707 | 0.9533 |
| 0.0412 | 4.16 | 2100 | 0.1948 | 0.9533 |
| 0.0338 | 4.26 | 2150 | 0.2039 | 0.9533 |
| 0.0796 | 4.36 | 2200 | 0.1797 | 0.9533 |
| 0.0727 | 4.46 | 2250 | 0.1986 | 0.9533 |
| 0.032 | 4.55 | 2300 | 0.1947 | 0.9467 |
| 0.0436 | 4.65 | 2350 | 0.1908 | 0.9467 |
| 0.0205 | 4.75 | 2400 | 0.1806 | 0.96 |
| 0.0326 | 4.85 | 2450 | 0.1835 | 0.96 |
| 0.0404 | 4.95 | 2500 | 0.1832 | 0.96 |
Framework versions
- Transformers 4.24.0
- Pytorch 1.13.0
- Datasets 2.7.1
- Tokenizers 0.13.2
