<!-- 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. -->
MT-legendary-capybara-96
This model is a fine-tuned version of toobiza/MT-ancient-spaceship-83 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.1572
- Loss Ce: 0.0000
- Loss Bbox: 0.0216
- Cardinality Error: 1.0
- Giou: 97.5514
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: 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: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Loss Ce | Loss Bbox | Cardinality Error | Giou |
|---|---|---|---|---|---|---|---|
| 0.2851 | 0.24 | 200 | 0.1903 | 0.0000 | 0.0263 | 1.0 | 97.0566 |
| 0.1809 | 0.48 | 400 | 0.1726 | 0.0000 | 0.0237 | 1.0 | 97.2974 |
| 0.1909 | 0.73 | 600 | 0.1923 | 0.0000 | 0.0268 | 1.0 | 97.0772 |
| 0.1808 | 0.97 | 800 | 0.1745 | 0.0000 | 0.0239 | 1.0 | 97.2598 |
| 0.169 | 1.21 | 1000 | 0.1774 | 0.0000 | 0.0245 | 1.0 | 97.2469 |
| 0.1916 | 1.45 | 1200 | 0.1800 | 0.0000 | 0.0249 | 1.0 | 97.2128 |
| 0.1511 | 1.69 | 1400 | 0.1810 | 0.0000 | 0.0251 | 1.0 | 97.2199 |
| 0.1205 | 1.93 | 1600 | 0.1811 | 0.0000 | 0.0251 | 1.0 | 97.2107 |
| 0.0905 | 2.18 | 1800 | 0.1816 | 0.0000 | 0.0252 | 1.0 | 97.2090 |
| 0.1175 | 2.42 | 2000 | 0.1789 | 0.0000 | 0.0247 | 1.0 | 97.2187 |
| 0.1781 | 2.66 | 2200 | 0.1713 | 0.0000 | 0.0236 | 1.0 | 97.3242 |
| 0.1751 | 2.9 | 2400 | 0.1886 | 0.0000 | 0.0261 | 1.0 | 97.0914 |
| 0.1084 | 3.14 | 2600 | 0.1692 | 0.0000 | 0.0232 | 1.0 | 97.3369 |
| 0.1171 | 3.39 | 2800 | 0.1570 | 0.0000 | 0.0216 | 1.0 | 97.5552 |
| 0.1191 | 3.63 | 3000 | 0.1859 | 0.0000 | 0.0259 | 1.0 | 97.1879 |
| 0.1515 | 3.87 | 3200 | 0.1598 | 0.0000 | 0.0221 | 1.0 | 97.5370 |
| 0.1529 | 4.11 | 3400 | 0.1750 | 0.0000 | 0.0240 | 1.0 | 97.2571 |
| 0.1169 | 4.35 | 3600 | 0.1627 | 0.0000 | 0.0224 | 1.0 | 97.4536 |
| 0.1433 | 4.59 | 3800 | 0.1764 | 0.0000 | 0.0244 | 1.0 | 97.2739 |
| 0.0873 | 4.84 | 4000 | 0.1536 | 0.0000 | 0.0209 | 1.0 | 97.5448 |
| 0.1176 | 5.08 | 4200 | 0.1545 | 0.0000 | 0.0212 | 1.0 | 97.5786 |
| 0.0921 | 5.32 | 4400 | 0.1580 | 0.0000 | 0.0216 | 1.0 | 97.5027 |
| 0.0894 | 5.56 | 4600 | 0.1579 | 0.0000 | 0.0216 | 1.0 | 97.5178 |
| 0.0843 | 5.8 | 4800 | 0.1604 | 0.0000 | 0.0220 | 1.0 | 97.4857 |
| 0.1446 | 6.05 | 5000 | 0.1692 | 0.0000 | 0.0233 | 1.0 | 97.3695 |
| 0.0929 | 6.29 | 5200 | 0.1723 | 0.0000 | 0.0238 | 1.0 | 97.3369 |
| 0.0831 | 6.53 | 5400 | 0.1638 | 0.0000 | 0.0225 | 1.0 | 97.4370 |
| 0.093 | 6.77 | 5600 | 0.1606 | 0.0000 | 0.0220 | 1.0 | 97.4782 |
| 0.0869 | 7.01 | 5800 | 0.1604 | 0.0000 | 0.0220 | 1.0 | 97.4893 |
| 0.1183 | 7.26 | 6000 | 0.1599 | 0.0000 | 0.0219 | 1.0 | 97.4886 |
| 0.0807 | 7.5 | 6200 | 0.1614 | 0.0000 | 0.0222 | 1.0 | 97.4926 |
| 0.0851 | 7.74 | 6400 | 0.1642 | 0.0000 | 0.0226 | 1.0 | 97.4411 |
| 0.1279 | 7.98 | 6600 | 0.1596 | 0.0000 | 0.0220 | 1.0 | 97.5193 |
| 0.0828 | 8.22 | 6800 | 0.1606 | 0.0000 | 0.0222 | 1.0 | 97.5183 |
| 0.0933 | 8.46 | 7000 | 0.1576 | 0.0000 | 0.0217 | 1.0 | 97.5506 |
| 0.085 | 8.71 | 7200 | 0.1584 | 0.0000 | 0.0218 | 1.0 | 97.5329 |
| 0.0736 | 8.95 | 7400 | 0.1564 | 0.0000 | 0.0215 | 1.0 | 97.5616 |
| 0.1001 | 9.19 | 7600 | 0.1581 | 0.0000 | 0.0217 | 1.0 | 97.5258 |
| 0.075 | 9.43 | 7800 | 0.1575 | 0.0000 | 0.0217 | 1.0 | 97.5435 |
| 0.0714 | 9.67 | 8000 | 0.1571 | 0.0000 | 0.0216 | 1.0 | 97.5487 |
| 0.0881 | 9.92 | 8200 | 0.1572 | 0.0000 | 0.0216 | 1.0 | 97.5514 |
Framework versions
- Transformers 4.33.2
- Pytorch 2.1.0+cu118
- Datasets 2.14.6
- Tokenizers 0.13.3
