<!-- 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_roBERTa_5E
This model is a fine-tuned version of roberta-base on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.2136
- Accuracy: 0.9667
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 |
---|---|---|---|---|
0.4665 | 0.1 | 50 | 0.2587 | 0.9333 |
0.245 | 0.2 | 100 | 0.1355 | 0.96 |
0.2079 | 0.3 | 150 | 0.1454 | 0.9533 |
0.2098 | 0.4 | 200 | 0.1809 | 0.9533 |
0.1637 | 0.5 | 250 | 0.2299 | 0.94 |
0.1869 | 0.59 | 300 | 0.1324 | 0.9667 |
0.2202 | 0.69 | 350 | 0.1786 | 0.9467 |
0.2084 | 0.79 | 400 | 0.1541 | 0.9533 |
0.148 | 0.89 | 450 | 0.1790 | 0.9533 |
0.1945 | 0.99 | 500 | 0.1168 | 0.9667 |
0.1648 | 1.09 | 550 | 0.1153 | 0.96 |
0.1099 | 1.19 | 600 | 0.1239 | 0.96 |
0.1238 | 1.29 | 650 | 0.1486 | 0.9533 |
0.1067 | 1.39 | 700 | 0.1195 | 0.96 |
0.1324 | 1.49 | 750 | 0.1134 | 0.96 |
0.1128 | 1.58 | 800 | 0.1180 | 0.9667 |
0.1406 | 1.68 | 850 | 0.2081 | 0.9533 |
0.1516 | 1.78 | 900 | 0.1987 | 0.9533 |
0.1537 | 1.88 | 950 | 0.1644 | 0.96 |
0.0957 | 1.98 | 1000 | 0.1660 | 0.96 |
0.0699 | 2.08 | 1050 | 0.2057 | 0.9533 |
0.1007 | 2.18 | 1100 | 0.2336 | 0.9533 |
0.0677 | 2.28 | 1150 | 0.2399 | 0.9467 |
0.059 | 2.38 | 1200 | 0.2331 | 0.96 |
0.1051 | 2.48 | 1250 | 0.1974 | 0.9533 |
0.0778 | 2.57 | 1300 | 0.2857 | 0.9467 |
0.1099 | 2.67 | 1350 | 0.2641 | 0.9533 |
0.0747 | 2.77 | 1400 | 0.2219 | 0.9533 |
0.0874 | 2.87 | 1450 | 0.2780 | 0.9533 |
0.0675 | 2.97 | 1500 | 0.1993 | 0.96 |
0.052 | 3.07 | 1550 | 0.1918 | 0.96 |
0.0214 | 3.17 | 1600 | 0.2410 | 0.96 |
0.0512 | 3.27 | 1650 | 0.2353 | 0.96 |
0.0548 | 3.37 | 1700 | 0.2722 | 0.9533 |
0.0554 | 3.47 | 1750 | 0.1593 | 0.9733 |
0.0742 | 3.56 | 1800 | 0.2568 | 0.96 |
0.064 | 3.66 | 1850 | 0.2358 | 0.96 |
0.052 | 3.76 | 1900 | 0.2161 | 0.9667 |
0.0349 | 3.86 | 1950 | 0.2497 | 0.96 |
0.0868 | 3.96 | 2000 | 0.1834 | 0.9667 |
0.0445 | 4.06 | 2050 | 0.2441 | 0.9533 |
0.0388 | 4.16 | 2100 | 0.2136 | 0.9667 |
0.0484 | 4.26 | 2150 | 0.2114 | 0.9667 |
0.0263 | 4.36 | 2200 | 0.2325 | 0.96 |
0.0409 | 4.46 | 2250 | 0.2454 | 0.9533 |
0.0324 | 4.55 | 2300 | 0.2105 | 0.9667 |
0.0295 | 4.65 | 2350 | 0.2118 | 0.9667 |
0.0372 | 4.75 | 2400 | 0.2005 | 0.9667 |
0.0294 | 4.85 | 2450 | 0.2057 | 0.9667 |
0.0354 | 4.95 | 2500 | 0.2136 | 0.9667 |
Framework versions
- Transformers 4.24.0
- Pytorch 1.13.0
- Datasets 2.7.1
- Tokenizers 0.13.2