| --- |
| library_name: transformers |
| license: apache-2.0 |
| base_model: deepvk/RuModernBERT-base |
| tags: |
| - generated_from_trainer |
| metrics: |
| - f1 |
| - precision |
| - recall |
| model-index: |
| - name: rumodernbert_ner_ft |
| results: [] |
| --- |
| |
| <!-- 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. --> |
|
|
| # rumodernbert_ner_ft |
|
|
| This model is a fine-tuned version of [deepvk/RuModernBERT-base](https://huggingface.co/deepvk/RuModernBERT-base) on an unknown dataset. |
| It achieves the following results on the evaluation set: |
| - Loss: 0.2121 |
| - F1: 0.8730 |
| - Precision: 0.8596 |
| - Recall: 0.8868 |
|
|
| ## 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: 2e-05 |
| - train_batch_size: 16 |
| - eval_batch_size: 32 |
| - seed: 42 |
| - gradient_accumulation_steps: 2 |
| - total_train_batch_size: 32 |
| - 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 |
| - lr_scheduler_warmup_steps: 0.1 |
| - num_epochs: 10 |
| - mixed_precision_training: Native AMP |
| |
| ### Training results |
| |
| | Training Loss | Epoch | Step | Validation Loss | F1 | Precision | Recall | |
| |:-------------:|:-----:|:----:|:---------------:|:------:|:---------:|:------:| |
| | 0.5866 | 0.32 | 200 | 0.2772 | 0.7188 | 0.6785 | 0.7640 | |
| | 0.4701 | 0.64 | 400 | 0.2204 | 0.7867 | 0.7535 | 0.8230 | |
| | 0.4275 | 0.96 | 600 | 0.1895 | 0.8191 | 0.7932 | 0.8467 | |
| | 0.3282 | 1.28 | 800 | 0.1970 | 0.8339 | 0.8155 | 0.8531 | |
| | 0.2545 | 1.6 | 1000 | 0.1765 | 0.8553 | 0.8427 | 0.8684 | |
| | 0.2703 | 1.92 | 1200 | 0.1528 | 0.8626 | 0.8512 | 0.8744 | |
| | 0.1658 | 2.24 | 1400 | 0.1862 | 0.8610 | 0.8476 | 0.8748 | |
| | 0.1571 | 2.56 | 1600 | 0.1882 | 0.8617 | 0.8464 | 0.8776 | |
| | 0.1604 | 2.88 | 1800 | 0.1715 | 0.8760 | 0.8638 | 0.8884 | |
| | 0.0633 | 3.2 | 2000 | 0.2327 | 0.8656 | 0.8443 | 0.8880 | |
| | 0.0925 | 3.52 | 2200 | 0.2032 | 0.8757 | 0.8641 | 0.8876 | |
| | 0.0730 | 3.84 | 2400 | 0.2121 | 0.8730 | 0.8596 | 0.8868 | |
| |
| |
| ### Framework versions |
| |
| - Transformers 5.1.0 |
| - Pytorch 2.10.0+cu128 |
| - Datasets 4.7.0 |
| - Tokenizers 0.22.2 |
| |
