<!-- 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-2e-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: 0.1327
- Precision: 0.7423
- Recall: 0.5632
- F1: 0.6405
- Accuracy: 0.9613
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: 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 |
|---|---|---|---|---|---|---|---|
| 0.165 | 1.0 | 1312 | 0.1422 | 0.6965 | 0.4508 | 0.5473 | 0.9530 |
| 0.1317 | 2.0 | 2624 | 0.1309 | 0.7152 | 0.5191 | 0.6016 | 0.9571 |
| 0.1113 | 3.0 | 3936 | 0.1311 | 0.7385 | 0.5341 | 0.6199 | 0.9593 |
| 0.0977 | 4.0 | 5248 | 0.1297 | 0.7474 | 0.5412 | 0.6278 | 0.9604 |
| 0.0862 | 5.0 | 6560 | 0.1327 | 0.7423 | 0.5632 | 0.6405 | 0.9613 |
Framework versions
- Transformers 4.30.2
- Pytorch 2.0.1+cu118
- Datasets 2.13.1
- Tokenizers 0.13.3
