metadata
license: mit
base_model: xlm-roberta-base
tags:
- generated_from_trainer
datasets:
- masakhaner2
metrics:
- f1
model-index:
- name: xlm-roberta-base-finetuned-wolof
results:
- task:
name: Token Classification
type: token-classification
dataset:
name: masakhaner2
type: masakhaner2
config: wol
split: validation
args: wol
metrics:
- name: F1
type: f1
value: 0.8514106583072101
xlm-roberta-base-finetuned-wolof
This model is a fine-tuned version of xlm-roberta-base on the masakhaner2 dataset. It achieves the following results on the evaluation set:
- Loss: 0.4671
- F1: 0.8514
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: 5.1193219561473124e-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: 7
Training results
Training Loss | Epoch | Step | Validation Loss | F1 |
---|---|---|---|---|
0.6236 | 1.0 | 739 | 0.3714 | 0.7072 |
0.296 | 2.0 | 1478 | 0.3902 | 0.7642 |
0.1785 | 3.0 | 2217 | 0.3680 | 0.7728 |
0.1222 | 4.0 | 2956 | 0.3825 | 0.8232 |
0.0727 | 5.0 | 3695 | 0.3973 | 0.8274 |
0.042 | 6.0 | 4434 | 0.5533 | 0.8460 |
0.0233 | 7.0 | 5173 | 0.4671 | 0.8514 |
Framework versions
- Transformers 4.33.0
- Pytorch 2.0.0
- Datasets 2.1.0
- Tokenizers 0.13.3