BERTModified-finetuned-wikitext-test
This model is a fine-tuned version of distilbert-base-uncased on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 18.8994
- Precision: 0.25
- Recall: 0.25
- F1: 0.25
- Accuracy: 0.25
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: 4
- eval_batch_size: 4
- 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 | Precision | Recall | F1 | Accuracy |
---|---|---|---|---|---|---|---|
19.9877 | 1.0 | 250 | 19.8070 | 0.0385 | 0.0385 | 0.0385 | 0.0385 |
15.4776 | 2.0 | 500 | 20.2930 | 0.0577 | 0.0577 | 0.0577 | 0.0577 |
13.1238 | 3.0 | 750 | 20.1112 | 0.0769 | 0.0769 | 0.0769 | 0.0769 |
11.1387 | 4.0 | 1000 | 19.9105 | 0.0897 | 0.0897 | 0.0897 | 0.0897 |
9.5317 | 5.0 | 1250 | 19.9108 | 0.1282 | 0.1282 | 0.1282 | 0.1282 |
8.037 | 6.0 | 1500 | 19.6093 | 0.1410 | 0.1410 | 0.1410 | 0.1410 |
6.7498 | 7.0 | 1750 | 19.1636 | 0.1474 | 0.1474 | 0.1474 | 0.1474 |
5.6472 | 8.0 | 2000 | 19.6709 | 0.1538 | 0.1538 | 0.1538 | 0.1538 |
4.6665 | 9.0 | 2250 | 19.2537 | 0.1667 | 0.1667 | 0.1667 | 0.1667 |
3.9107 | 10.0 | 2500 | 19.1982 | 0.1474 | 0.1474 | 0.1474 | 0.1474 |
3.1874 | 11.0 | 2750 | 18.9938 | 0.1731 | 0.1731 | 0.1731 | 0.1731 |
2.5846 | 12.0 | 3000 | 18.7462 | 0.2115 | 0.2115 | 0.2115 | 0.2115 |
2.1464 | 13.0 | 3250 | 19.0017 | 0.1667 | 0.1667 | 0.1667 | 0.1667 |
1.7521 | 14.0 | 3500 | 18.4513 | 0.1859 | 0.1859 | 0.1859 | 0.1859 |
1.4561 | 15.0 | 3750 | 18.7532 | 0.2051 | 0.2051 | 0.2051 | 0.2051 |
1.2254 | 16.0 | 4000 | 18.3970 | 0.2179 | 0.2179 | 0.2179 | 0.2179 |
1.0416 | 17.0 | 4250 | 18.9764 | 0.1859 | 0.1859 | 0.1859 | 0.1859 |
0.8923 | 18.0 | 4500 | 18.3271 | 0.2244 | 0.2244 | 0.2244 | 0.2244 |
0.7803 | 19.0 | 4750 | 18.5893 | 0.2436 | 0.2436 | 0.2436 | 0.2436 |
0.6839 | 20.0 | 5000 | 18.3505 | 0.2051 | 0.2051 | 0.2051 | 0.2051 |
0.6175 | 21.0 | 5250 | 18.6798 | 0.2051 | 0.2051 | 0.2051 | 0.2051 |
0.5491 | 22.0 | 5500 | 18.7426 | 0.2115 | 0.2115 | 0.2115 | 0.2115 |
0.4952 | 23.0 | 5750 | 18.3955 | 0.2179 | 0.2179 | 0.2179 | 0.2179 |
0.4441 | 24.0 | 6000 | 18.5502 | 0.2564 | 0.2564 | 0.2564 | 0.2564 |
0.4047 | 25.0 | 6250 | 18.9599 | 0.2244 | 0.2244 | 0.2244 | 0.2244 |
0.3768 | 26.0 | 6500 | 18.8141 | 0.2308 | 0.2308 | 0.2308 | 0.2308 |
0.3435 | 27.0 | 6750 | 18.9732 | 0.2436 | 0.2436 | 0.2436 | 0.2436 |
0.3164 | 28.0 | 7000 | 18.9216 | 0.2372 | 0.2372 | 0.2372 | 0.2372 |
0.2954 | 29.0 | 7250 | 18.6152 | 0.1987 | 0.1987 | 0.1987 | 0.1987 |
0.2736 | 30.0 | 7500 | 18.6001 | 0.25 | 0.25 | 0.25 | 0.25 |
0.2491 | 31.0 | 7750 | 19.1374 | 0.2436 | 0.2436 | 0.2436 | 0.2436 |
0.2359 | 32.0 | 8000 | 18.8624 | 0.25 | 0.25 | 0.25 | 0.25 |
0.2222 | 33.0 | 8250 | 18.3201 | 0.2308 | 0.2308 | 0.2308 | 0.2308 |
0.212 | 34.0 | 8500 | 18.7708 | 0.2179 | 0.2179 | 0.2179 | 0.2179 |
0.1864 | 35.0 | 8750 | 18.8994 | 0.2372 | 0.2372 | 0.2372 | 0.2372 |
0.1771 | 36.0 | 9000 | 18.3130 | 0.2308 | 0.2308 | 0.2308 | 0.2308 |
0.1703 | 37.0 | 9250 | 18.6183 | 0.2436 | 0.2436 | 0.2436 | 0.2436 |
0.1554 | 38.0 | 9500 | 18.8593 | 0.2372 | 0.2372 | 0.2372 | 0.2372 |
0.1469 | 39.0 | 9750 | 18.8936 | 0.2628 | 0.2628 | 0.2628 | 0.2628 |
0.1407 | 40.0 | 10000 | 18.9002 | 0.2372 | 0.2372 | 0.2372 | 0.2372 |
0.1328 | 41.0 | 10250 | 19.1827 | 0.2564 | 0.2564 | 0.2564 | 0.2564 |
0.1297 | 42.0 | 10500 | 18.5465 | 0.25 | 0.25 | 0.25 | 0.25 |
0.1226 | 43.0 | 10750 | 18.9125 | 0.2308 | 0.2308 | 0.2308 | 0.2308 |
0.1218 | 44.0 | 11000 | 19.0831 | 0.2308 | 0.2308 | 0.2308 | 0.2308 |
0.1136 | 45.0 | 11250 | 18.7969 | 0.2372 | 0.2372 | 0.2372 | 0.2372 |
0.1075 | 46.0 | 11500 | 18.7629 | 0.25 | 0.25 | 0.25 | 0.25 |
0.1044 | 47.0 | 11750 | 18.9700 | 0.2115 | 0.2115 | 0.2115 | 0.2115 |
0.1042 | 48.0 | 12000 | 18.7211 | 0.2628 | 0.2628 | 0.2628 | 0.2628 |
0.1008 | 49.0 | 12250 | 18.9104 | 0.2244 | 0.2244 | 0.2244 | 0.2244 |
0.1014 | 50.0 | 12500 | 18.7892 | 0.25 | 0.25 | 0.25 | 0.25 |
Framework versions
- Transformers 4.24.0
- Pytorch 1.13.0
- Datasets 2.6.1
- Tokenizers 0.13.2