<!-- 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. -->
afro-xlmr-base-hausa-5e-5
This model is a fine-tuned version of Davlan/afro-xlmr-base on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.1512
- Precision: 0.7391
- Recall: 0.5807
- F1: 0.6504
- Accuracy: 0.9616
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: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|---|---|---|---|---|---|---|---|
| 0.1604 | 1.0 | 1312 | 0.1395 | 0.6845 | 0.4906 | 0.5716 | 0.9535 |
| 0.1221 | 2.0 | 2624 | 0.1261 | 0.7140 | 0.5440 | 0.6175 | 0.9582 |
| 0.0939 | 3.0 | 3936 | 0.1311 | 0.7433 | 0.5693 | 0.6448 | 0.9610 |
| 0.0723 | 4.0 | 5248 | 0.1419 | 0.7508 | 0.5583 | 0.6404 | 0.9613 |
| 0.0557 | 5.0 | 6560 | 0.1512 | 0.7391 | 0.5807 | 0.6504 | 0.9616 |
Framework versions
- Transformers 4.30.2
- Pytorch 2.0.1+cu118
- Datasets 2.13.1
- Tokenizers 0.13.3
