metadata
base_model: aubmindlab/bert-base-arabertv2
tags:
- generated_from_trainer
metrics:
- accuracy
- precision
- recall
- f1
model-index:
- name: arabertv2-fully-supervised-arabic-propaganda
results: []
arabertv2-fully-supervised-arabic-propaganda
This model is a fine-tuned version of aubmindlab/bert-base-arabertv2 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.3894
- Accuracy: 0.9262
- Precision: 0.6042
- Recall: 0.7073
- F1: 0.6517
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: 64
- eval_batch_size: 64
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 5
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 |
|---|---|---|---|---|---|---|---|
| 0.4919 | 1.0 | 20 | 0.5087 | 0.8381 | 0.3516 | 0.7805 | 0.4848 |
| 0.3633 | 2.0 | 40 | 0.4010 | 0.8333 | 0.3474 | 0.8049 | 0.4853 |
| 0.2017 | 3.0 | 60 | 0.3635 | 0.9 | 0.4918 | 0.7317 | 0.5882 |
| 0.3071 | 4.0 | 80 | 0.3981 | 0.9333 | 0.6444 | 0.7073 | 0.6744 |
| 0.145 | 5.0 | 100 | 0.3894 | 0.9262 | 0.6042 | 0.7073 | 0.6517 |
Framework versions
- Transformers 4.35.0
- Pytorch 2.1.0+cu121
- Datasets 2.14.6
- Tokenizers 0.14.1