ossbert-morph-v2.1
This model is a fine-tuned version of AlexeySorokin/ossbert-onc-unlab-from_multilingual-bs64-5epochs on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.3005
- Accuracy: 96.1278
- Sentence accuracy: 63.3028
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: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- num_epochs: 25
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Sentence accuracy |
|---|---|---|---|---|---|
| 0.9799 | 1.0 | 546 | 0.3471 | 92.2180 | 41.6514 |
| 0.3099 | 2.0 | 1092 | 0.2477 | 94.3358 | 52.8440 |
| 0.1942 | 3.0 | 1638 | 0.2150 | 94.7243 | 55.5963 |
| 0.1367 | 4.0 | 2184 | 0.2289 | 94.8872 | 54.6789 |
| 0.1013 | 5.0 | 2730 | 0.2131 | 95.5890 | 58.1651 |
| 0.0818 | 6.0 | 3276 | 0.2122 | 95.6516 | 58.3486 |
| 0.0621 | 7.0 | 3822 | 0.2380 | 95.6642 | 60.1835 |
| 0.0457 | 8.0 | 4368 | 0.2346 | 95.6642 | 59.0826 |
| 0.0381 | 9.0 | 4914 | 0.2426 | 95.7393 | 59.0826 |
| 0.0273 | 10.0 | 5460 | 0.2629 | 95.7895 | 60.5505 |
| 0.0163 | 11.0 | 6006 | 0.2672 | 95.7018 | 59.0826 |
| 0.0127 | 12.0 | 6552 | 0.2782 | 95.8020 | 60.7339 |
| 0.0095 | 13.0 | 7098 | 0.2812 | 95.7268 | 60.9174 |
| 0.0086 | 14.0 | 7644 | 0.2731 | 96.1529 | 63.4862 |
| 0.0077 | 15.0 | 8190 | 0.2939 | 95.8396 | 60.0 |
| 0.0065 | 16.0 | 8736 | 0.2985 | 95.9900 | 62.0183 |
| 0.0055 | 17.0 | 9282 | 0.2955 | 95.9774 | 61.6514 |
| 0.0029 | 18.0 | 9828 | 0.2982 | 96.0526 | 61.4679 |
| 0.0034 | 19.0 | 10374 | 0.3005 | 96.1278 | 63.3028 |
Framework versions
- Transformers 4.57.3
- Pytorch 2.11.0+cu130
- Datasets 4.0.0
- Tokenizers 0.22.2