This model is a fine-tuned version of masakhane/afroxlmr-large-ner-masakhaner-1.0_2.0 on the Beijuka/Multilingual_PII_NER_dataset dataset. It achieves the following results on the evaluation set:

Loss: 0.0833
Precision: 0.9154
Recall: 0.9276
F1: 0.9214
Accuracy: 0.9827

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_FUSED 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	301	0.1004	0.9228	0.8951	0.9088	0.9732
0.1466	2.0	602	0.1025	0.8953	0.9227	0.9088	0.9761
0.1466	3.0	903	0.0914	0.9128	0.9287	0.9207	0.9781
0.069	4.0	1204	0.1235	0.8805	0.9359	0.9073	0.9752
0.0477	5.0	1505	0.1069	0.9179	0.9179	0.9179	0.9762
0.0477	6.0	1806	0.1166	0.8885	0.9449	0.9158	0.9764

Framework versions

Transformers 4.56.0
Pytorch 2.8.0.dev20250319+cu128
Datasets 4.0.0
Tokenizers 0.22.0

Beijuka/afroxlmr-large-ner-masakhaner-1.0_2.0-kanuri-ner-v1

multilingual-masakhane/afroxlmr-large-ner-masakhaner-1.0_2.0-kanuri-ner-v1

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

Beijuka/
afroxlmr-large-ner-masakhaner-1.0_2.0-kanuri-ner-v1