metadata
license: mit
base_model: urduhack/roberta-urdu-small
tags:
- generated_from_trainer
datasets:
- massive
model-index:
- name: UrduIntentClassification
results: []
widget:
- text: جنگ کی کيا نئی خبر ہے
example_title: News
- text: ہر ہفتے کے دن آٹھ کے لیے الارم سیٹ کریں
example_title: Alarm
- text: میں mcdonald کو ایک شکایت ٹویٹ کرنا چاہتا ہوں
example_title: Social
UrduIntentClassification
This model is a fine-tuned version of urduhack/roberta-urdu-small on the massive dataset. It achieves the following results on the evaluation set:
- Loss: 0.7482
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: 5e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- num_epochs: 8
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 1.3314 | 1.0 | 720 | 1.0549 |
| 0.7468 | 2.0 | 1440 | 0.7997 |
| 0.4117 | 3.0 | 2160 | 0.7482 |
| 0.3267 | 4.0 | 2880 | 0.8247 |
| 0.2292 | 5.0 | 3600 | 0.9014 |
| 0.0356 | 6.0 | 4320 | 0.9446 |
| 0.0123 | 7.0 | 5040 | 0.9757 |
| 0.0208 | 8.0 | 5760 | 0.9854 |
Framework versions
- Transformers 4.32.1
- Pytorch 2.0.0
- Datasets 2.1.0
- Tokenizers 0.13.3