| --- |
| license: mit |
| tags: |
| - generated_from_trainer |
| datasets: |
| - lg-ner |
| metrics: |
| - precision |
| - recall |
| - f1 |
| - accuracy |
| model-index: |
| - name: luganda-ner-v6 |
| results: |
| - task: |
| name: Token Classification |
| type: token-classification |
| dataset: |
| name: lg-ner |
| type: lg-ner |
| config: lug |
| split: test |
| args: lug |
| metrics: |
| - name: Precision |
| type: precision |
| value: 0.8241451500348919 |
| - name: Recall |
| type: recall |
| value: 0.7931497649429147 |
| - name: F1 |
| type: f1 |
| value: 0.8083504449007528 |
| - name: Accuracy |
| type: accuracy |
| value: 0.9525918396979817 |
| --- |
| |
| <!-- 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. --> |
|
|
| # luganda-ner-v6 |
|
|
| This model is a fine-tuned version of [Davlan/afro-xlmr-base](https://huggingface.co/Davlan/afro-xlmr-base) on the lg-ner dataset. |
| It achieves the following results on the evaluation set: |
| - Loss: 0.2417 |
| - Precision: 0.8241 |
| - Recall: 0.7931 |
| - F1: 0.8084 |
| - Accuracy: 0.9526 |
|
|
| ## 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: 8 |
| - eval_batch_size: 8 |
| - 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 | Precision | Recall | F1 | Accuracy | |
| |:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:| |
| | No log | 1.0 | 261 | 0.4290 | 0.5281 | 0.3096 | 0.3903 | 0.8864 | |
| | 0.5483 | 2.0 | 522 | 0.2873 | 0.7307 | 0.5776 | 0.6452 | 0.9216 | |
| | 0.5483 | 3.0 | 783 | 0.2482 | 0.7745 | 0.6783 | 0.7232 | 0.9334 | |
| | 0.1931 | 4.0 | 1044 | 0.2472 | 0.7671 | 0.6991 | 0.7316 | 0.9360 | |
| | 0.1931 | 5.0 | 1305 | 0.2425 | 0.8053 | 0.7388 | 0.7706 | 0.9433 | |
| | 0.1016 | 6.0 | 1566 | 0.2157 | 0.8253 | 0.7710 | 0.7972 | 0.9490 | |
| | 0.1016 | 7.0 | 1827 | 0.2332 | 0.8161 | 0.7717 | 0.7932 | 0.9501 | |
| | 0.0654 | 8.0 | 2088 | 0.2375 | 0.8312 | 0.7804 | 0.8050 | 0.9514 | |
| | 0.0654 | 9.0 | 2349 | 0.2367 | 0.8309 | 0.7884 | 0.8091 | 0.9528 | |
| | 0.047 | 10.0 | 2610 | 0.2417 | 0.8241 | 0.7931 | 0.8084 | 0.9526 | |
|
|
|
|
| ### Framework versions |
|
|
| - Transformers 4.27.4 |
| - Pytorch 1.13.1+cu116 |
| - Datasets 2.11.0 |
| - Tokenizers 0.13.2 |
|
|
