|
--- |
|
license: cc-by-4.0 |
|
datasets: |
|
- Sankalp-Bahad/NER-Dataset |
|
language: |
|
- ur |
|
metrics: |
|
- f1 |
|
library_name: transformers |
|
pipeline_tag: token-classification |
|
--- |
|
|
|
# Monolingual-Urdu-NER-Model |
|
|
|
This model is developed by fine-tuning XLM-Roberta-Base on the NER-Dataset. |
|
|
|
## Training procedure |
|
|
|
### Training hyperparameters |
|
|
|
The following hyperparameters were used during training: |
|
- learning_rate: 5e-05 |
|
- train_batch_size: 4 |
|
- eval_batch_size: 4 |
|
- optimizer: Adam |
|
- lr_scheduler_type: linear |
|
- num_epochs: 10.0 |
|
|
|
### Framework versions |
|
|
|
- Transformers 4.38.2 |
|
- Pytorch 1.9.1 |
|
- Datasets 2.14.6 |
|
- Tokenizers 0.15.0 |
