---
library_name: transformers
language:
- lg
license: apache-2.0
base_model: google-bert/bert-base-multilingual-cased
tags:
- named-entity-recognition
- luganda
- african-languages
- pii-detection
- token-classification
- generated_from_trainer
datasets:
- Beijuka/Luganda_Monolingual_PII_dataset
metrics:
- precision
- recall
- f1
- accuracy
model-index:
- name: luganda-ner-bert-v7
  results:
  - task:
      name: Token Classification
      type: token-classification
    dataset:
      name: Beijuka/Luganda_Monolingual_PII_dataset
      type: Beijuka/Luganda_Monolingual_PII_dataset
      args: 'split: train+validation+test'
    metrics:
    - name: Precision
      type: precision
      value: 0.8317757009345794
    - name: Recall
      type: recall
      value: 0.7954319761668321
    - name: F1
      type: f1
      value: 0.8131979695431472
    - name: Accuracy
      type: accuracy
      value: 0.9486124353891705
---

<!-- 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-bert-v7

This model is a fine-tuned version of [google-bert/bert-base-multilingual-cased](https://huggingface.co/google-bert/bert-base-multilingual-cased) on the Beijuka/Luganda_Monolingual_PII_dataset dataset.
It achieves the following results on the evaluation set:
- Loss: 0.3956
- Precision: 0.8318
- Recall: 0.7954
- F1: 0.8132
- Accuracy: 0.9486

## 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 with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- num_epochs: 20

### Training results

| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1     | Accuracy |
|:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
| No log        | 1.0   | 261  | 0.4093          | 0.6488    | 0.4697 | 0.5449 | 0.8871   |
| 0.5031        | 2.0   | 522  | 0.3278          | 0.6709    | 0.6842 | 0.6775 | 0.9099   |
| 0.5031        | 3.0   | 783  | 0.2721          | 0.7816    | 0.7071 | 0.7424 | 0.9317   |
| 0.1642        | 4.0   | 1044 | 0.2540          | 0.7776    | 0.7776 | 0.7776 | 0.9399   |
| 0.1642        | 5.0   | 1305 | 0.3182          | 0.7473    | 0.7607 | 0.7539 | 0.9280   |
| 0.0775        | 6.0   | 1566 | 0.3038          | 0.7843    | 0.7944 | 0.7893 | 0.9409   |
| 0.0775        | 7.0   | 1827 | 0.3407          | 0.8504    | 0.7567 | 0.8008 | 0.9400   |
| 0.0362        | 8.0   | 2088 | 0.3267          | 0.7886    | 0.8004 | 0.7945 | 0.9429   |
| 0.0362        | 9.0   | 2349 | 0.3229          | 0.8316    | 0.7845 | 0.8074 | 0.9499   |
| 0.0201        | 10.0  | 2610 | 0.3521          | 0.8475    | 0.7726 | 0.8083 | 0.9476   |
| 0.0201        | 11.0  | 2871 | 0.3629          | 0.8126    | 0.8054 | 0.8090 | 0.9468   |
| 0.0095        | 12.0  | 3132 | 0.3817          | 0.8012    | 0.8083 | 0.8047 | 0.9466   |
| 0.0095        | 13.0  | 3393 | 0.3956          | 0.8318    | 0.7954 | 0.8132 | 0.9486   |
| 0.0057        | 14.0  | 3654 | 0.3798          | 0.8219    | 0.7974 | 0.8095 | 0.9458   |
| 0.0057        | 15.0  | 3915 | 0.4196          | 0.8166    | 0.7825 | 0.7992 | 0.9445   |


### Framework versions

- Transformers 4.53.0
- Pytorch 2.6.0+cu124
- Datasets 3.6.0
- Tokenizers 0.21.2