---
license: mit
tags:
- generated_from_trainer
datasets:
- lg-ner
metrics:
- precision
- recall
- f1
- accuracy
model-index:
- name: luganda-ner-v4
  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: 0.7540871934604905
    - name: Recall
      type: recall
      value: 0.7454545454545455
    - name: F1
      type: f1
      value: 0.7497460209955976
    - name: Accuracy
      type: accuracy
      value: 0.9360226606759132
---

<!-- 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-v4

This model is a fine-tuned version of [microsoft/deberta-v3-base](https://huggingface.co/microsoft/deberta-v3-base) on the lg-ner dataset.
It achieves the following results on the evaluation set:
- Loss: 0.3024
- Precision: 0.7541
- Recall: 0.7455
- F1: 0.7497
- Accuracy: 0.9360

## 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  | 0.4811          | 0.5366    | 0.2768 | 0.3652 | 0.8752   |
| 0.5133        | 2.0   | 522  | 0.3632          | 0.6560    | 0.5380 | 0.5912 | 0.9021   |
| 0.5133        | 3.0   | 783  | 0.3104          | 0.7069    | 0.5993 | 0.6487 | 0.9207   |
| 0.2592        | 4.0   | 1044 | 0.3339          | 0.7494    | 0.6303 | 0.6847 | 0.9269   |
| 0.2592        | 5.0   | 1305 | 0.3153          | 0.7513    | 0.6593 | 0.7023 | 0.9318   |
| 0.167         | 6.0   | 1566 | 0.3071          | 0.7190    | 0.7219 | 0.7204 | 0.9291   |
| 0.167         | 7.0   | 1827 | 0.3072          | 0.7955    | 0.7071 | 0.7487 | 0.9360   |
| 0.1191        | 8.0   | 2088 | 0.3133          | 0.7505    | 0.7455 | 0.7480 | 0.9339   |
| 0.1191        | 9.0   | 2349 | 0.3132          | 0.7510    | 0.7394 | 0.7452 | 0.9349   |
| 0.092         | 10.0  | 2610 | 0.3024          | 0.7541    | 0.7455 | 0.7497 | 0.9360   |


### Framework versions

- Transformers 4.27.4
- Pytorch 1.13.1+cu116
- Datasets 2.11.0
- Tokenizers 0.13.2