nerel-bio-RuBioBERT-base
This model is a fine-tuned version of alexyalunin/RuBioBERT on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.6187
- Precision: 0.8107
- Recall: 0.8157
- F1: 0.8132
- Accuracy: 0.8721
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: 64
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|---|---|---|---|---|---|---|---|
| No log | 1.0 | 153 | 0.8937 | 0.7363 | 0.7078 | 0.7218 | 0.7998 |
| No log | 2.0 | 306 | 0.5753 | 0.7907 | 0.7798 | 0.7852 | 0.8527 |
| No log | 3.0 | 459 | 0.5308 | 0.7893 | 0.7856 | 0.7875 | 0.8559 |
| 0.9025 | 4.0 | 612 | 0.5477 | 0.7956 | 0.7962 | 0.7959 | 0.8618 |
| 0.9025 | 5.0 | 765 | 0.5732 | 0.8011 | 0.8037 | 0.8024 | 0.8646 |
| 0.9025 | 6.0 | 918 | 0.5551 | 0.8085 | 0.8082 | 0.8084 | 0.8693 |
| 0.133 | 7.0 | 1071 | 0.5813 | 0.8082 | 0.8082 | 0.8082 | 0.8701 |
| 0.133 | 8.0 | 1224 | 0.6006 | 0.8053 | 0.8131 | 0.8092 | 0.8687 |
| 0.133 | 9.0 | 1377 | 0.6183 | 0.8104 | 0.8159 | 0.8131 | 0.8723 |
| 0.0406 | 10.0 | 1530 | 0.6187 | 0.8107 | 0.8157 | 0.8132 | 0.8721 |
Framework versions
- Transformers 4.37.2
- Pytorch 2.1.0+cu121
- Datasets 2.17.1
- Tokenizers 0.15.2