metadata
tags:
- generated_from_trainer
metrics:
- accuracy
- precision
- recall
model-index:
- name: AraElectra-finetuned-fnd
results: []
AraElectra-finetuned-fnd
This model is a fine-tuned version of aubmindlab/araelectra-base-discriminator on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.6049
- Macro F1: 0.7672
- Accuracy: 0.7733
- Precision: 0.7667
- Recall: 0.7678
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: 25
- gradient_accumulation_steps: 2
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 4
Training results
| Training Loss | Epoch | Step | Validation Loss | Macro F1 | Accuracy | Precision | Recall |
|---|---|---|---|---|---|---|---|
| 0.5274 | 1.0 | 1597 | 0.4950 | 0.7459 | 0.7574 | 0.7526 | 0.7426 |
| 0.432 | 2.0 | 3194 | 0.4829 | 0.7558 | 0.7687 | 0.7669 | 0.7513 |
| 0.3366 | 3.0 | 4791 | 0.5234 | 0.7591 | 0.7629 | 0.7576 | 0.7632 |
| 0.2761 | 4.0 | 6388 | 0.6049 | 0.7672 | 0.7733 | 0.7667 | 0.7678 |
Framework versions
- Transformers 4.20.1
- Pytorch 1.11.0
- Datasets 2.1.0
- Tokenizers 0.12.1