metadata

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: 0.8368200836820083
          - name: Recall
            type: recall
            value: 0.7944389275074478
          - name: F1
            type: f1
            value: 0.815078960774325
          - name: Accuracy
            type: accuracy
            value: 0.9449490640839062

luganda-ner-bert-v7

This model is a fine-tuned version of google-bert/bert-base-multilingual-cased on the Beijuka/Luganda_Monolingual_PII_dataset dataset. It achieves the following results on the evaluation set:

Loss: 0.3699
Precision: 0.8368
Recall: 0.7944
F1: 0.8151
Accuracy: 0.9449

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: 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	0.5128	0.5595	0.2661	0.3607	0.8665
0.6082	2.0	522	0.3601	0.5826	0.6127	0.5973	0.9005
0.6082	3.0	783	0.2763	0.7691	0.6584	0.7095	0.9262
0.2267	4.0	1044	0.2551	0.7732	0.7279	0.7499	0.9344
0.2267	5.0	1305	0.2874	0.7308	0.7468	0.7387	0.9283
0.1155	6.0	1566	0.2874	0.7549	0.7646	0.7597	0.9364
0.1155	7.0	1827	0.2852	0.8069	0.7716	0.7888	0.9455
0.0605	8.0	2088	0.3063	0.7820	0.7944	0.7882	0.9420
0.0605	9.0	2349	0.3227	0.8022	0.7974	0.7998	0.9427
0.0345	10.0	2610	0.3288	0.8026	0.8034	0.8030	0.9408
0.0345	11.0	2871	0.3661	0.8010	0.8232	0.8119	0.9410
0.0182	12.0	3132	0.3699	0.8368	0.7944	0.8151	0.9449
0.0182	13.0	3393	0.3683	0.7958	0.7974	0.7966	0.9436
0.0133	14.0	3654	0.3811	0.8008	0.8064	0.8036	0.9422

Framework versions

Transformers 4.53.0
Pytorch 2.6.0+cu124
Datasets 3.6.0
Tokenizers 0.21.2