afro-xlmr-base-afr-finetuned-augmentation-LUNAR-afr-MICRO
This model is a fine-tuned version of sercetexam9/afro-xlmr-base-afr-finetuned-augmentation-LUNAR on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.0857
- F1: 0.8947
- Roc Auc: 0.9286
- Accuracy: 0.8721
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: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 100
- num_epochs: 20
Training results
| Training Loss | Epoch | Step | Validation Loss | F1 | Roc Auc | Accuracy |
|---|---|---|---|---|---|---|
| 0.113 | 1.0 | 120 | 0.0857 | 0.8947 | 0.9286 | 0.8721 |
| 0.0985 | 2.0 | 240 | 0.1054 | 0.8563 | 0.9099 | 0.8302 |
| 0.0656 | 3.0 | 360 | 0.1019 | 0.8876 | 0.9403 | 0.8679 |
| 0.045 | 4.0 | 480 | 0.1354 | 0.8524 | 0.9357 | 0.8155 |
| 0.0409 | 5.0 | 600 | 0.1110 | 0.8698 | 0.9369 | 0.8407 |
Framework versions
- Transformers 4.45.1
- Pytorch 2.4.0
- Datasets 3.0.1
- Tokenizers 0.20.0