|
--- |
|
license: mit |
|
tags: |
|
- generated_from_trainer |
|
datasets: |
|
- lg-ner |
|
metrics: |
|
- precision |
|
- recall |
|
- f1 |
|
- accuracy |
|
model-index: |
|
- name: luganda-ner-v1 |
|
results: |
|
- task: |
|
name: Token Classification |
|
type: token-classification |
|
dataset: |
|
name: lg-ner |
|
type: lg-ner |
|
config: lug |
|
split: train |
|
args: lug |
|
metrics: |
|
- name: Precision |
|
type: precision |
|
value: 0.4158878504672897 |
|
- name: Recall |
|
type: recall |
|
value: 0.5028248587570622 |
|
- name: F1 |
|
type: f1 |
|
value: 0.45524296675191817 |
|
- name: Accuracy |
|
type: accuracy |
|
value: 0.8060836501901141 |
|
--- |
|
|
|
<!-- 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-v1 |
|
|
|
This model is a fine-tuned version of [xlm-roberta-base](https://huggingface.co/xlm-roberta-base) on the lg-ner dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.7681 |
|
- Precision: 0.4159 |
|
- Recall: 0.5028 |
|
- F1: 0.4552 |
|
- Accuracy: 0.8061 |
|
|
|
## 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 | 25 | 0.9702 | 0.2686 | 0.3672 | 0.3103 | 0.7240 | |
|
| No log | 2.0 | 50 | 0.8977 | 0.2702 | 0.3785 | 0.3153 | 0.7468 | |
|
| No log | 3.0 | 75 | 0.8785 | 0.2517 | 0.4124 | 0.3126 | 0.7551 | |
|
| No log | 4.0 | 100 | 0.8608 | 0.2927 | 0.4746 | 0.3621 | 0.7567 | |
|
| No log | 5.0 | 125 | 0.7859 | 0.4053 | 0.4350 | 0.4196 | 0.7909 | |
|
| No log | 6.0 | 150 | 0.7728 | 0.4010 | 0.4350 | 0.4173 | 0.7901 | |
|
| No log | 7.0 | 175 | 0.7647 | 0.4118 | 0.4746 | 0.4409 | 0.7932 | |
|
| No log | 8.0 | 200 | 0.7800 | 0.3929 | 0.4972 | 0.4389 | 0.7985 | |
|
| No log | 9.0 | 225 | 0.7706 | 0.4211 | 0.4972 | 0.4560 | 0.8053 | |
|
| No log | 10.0 | 250 | 0.7681 | 0.4159 | 0.5028 | 0.4552 | 0.8061 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.24.0 |
|
- Pytorch 1.12.1+cu113 |
|
- Datasets 2.7.1 |
|
- Tokenizers 0.13.2 |
|
|
