metadata

license: mit
tags:
  - generated_from_trainer
datasets:
  - lg-ner
metrics:
  - precision
  - recall
  - f1
  - accuracy
model-index:
  - name: luganda-ner-v3
    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: 0.8141289437585734
          - name: Recall
            type: recall
            value: 0.7971793149764943
          - name: F1
            type: f1
            value: 0.8055649813369528
          - name: Accuracy
            type: accuracy
            value: 0.952700740525628

luganda-ner-v3

This model is a fine-tuned version of xlm-roberta-large on the lg-ner dataset. It achieves the following results on the evaluation set:

Loss: 0.2295
Precision: 0.8141
Recall: 0.7972
F1: 0.8056
Accuracy: 0.9527

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: 6

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	1.0	261	0.4226	0.6273	0.3606	0.4580	0.8928
0.5572	2.0	522	0.2835	0.7720	0.6185	0.6868	0.9219
0.5572	3.0	783	0.2740	0.7579	0.7401	0.7489	0.9311
0.1745	4.0	1044	0.2423	0.7895	0.7683	0.7788	0.9399
0.1745	5.0	1305	0.2273	0.8048	0.7945	0.7996	0.9498
0.086	6.0	1566	0.2295	0.8141	0.7972	0.8056	0.9527

Framework versions

Transformers 4.27.4
Pytorch 1.13.1+cu116
Datasets 2.11.0
Tokenizers 0.13.2