metadata
base_model: aubmindlab/bert-base-arabertv02-twitter
tags:
- generated_from_trainer
metrics:
- accuracy
- precision
- recall
- f1
model-index:
- name: arabert-weakly-supervised-arabic-propaganda
results: []
arabert-weakly-supervised-arabic-propaganda
This model is a fine-tuned version of aubmindlab/bert-base-arabertv02-twitter on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.3223
- Accuracy: 0.8389
- Precision: 0.7865
- Recall: 0.7764
- F1: 0.7814
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 |
|---|---|---|---|---|---|---|---|
| 0.3758 | 1.0 | 2272 | 0.3615 | 0.8193 | 0.7950 | 0.6909 | 0.7393 |
| 0.3421 | 2.0 | 4544 | 0.3431 | 0.8285 | 0.7523 | 0.8016 | 0.7762 |
| 0.3447 | 3.0 | 6816 | 0.3389 | 0.8305 | 0.7933 | 0.7345 | 0.7628 |
| 0.3229 | 4.0 | 9088 | 0.3297 | 0.8352 | 0.7725 | 0.7877 | 0.7800 |
| 0.3176 | 5.0 | 11360 | 0.3223 | 0.8389 | 0.7865 | 0.7764 | 0.7814 |
Framework versions
- Transformers 4.35.0
- Pytorch 2.1.0+cu121
- Datasets 2.14.6
- Tokenizers 0.14.1