metadata
library_name: transformers
base_model: whisper-base-ar-quran
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: whisper-base-ar-quran-ft-hijaiyah
results: []
whisper-base-ar-quran-ft-hijaiyah
This model is a fine-tuned version of whisper-base-ar-quran on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.1638
- Accuracy: 0.9802
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: 5e-05
- train_batch_size: 2
- eval_batch_size: 2
- seed: 42
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 10
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 0.3043 | 1.0 | 2266 | 0.7745 | 0.8075 |
| 0.9065 | 2.0 | 4532 | 0.7729 | 0.8671 |
| 0.7332 | 3.0 | 6798 | 0.6099 | 0.9028 |
| 0.2412 | 4.0 | 9064 | 0.4158 | 0.9345 |
| 0.6118 | 5.0 | 11330 | 0.4587 | 0.9385 |
| 0.0001 | 6.0 | 13596 | 0.2835 | 0.9603 |
| 0.0003 | 7.0 | 15862 | 0.2181 | 0.9643 |
| 0.0001 | 8.0 | 18128 | 0.2006 | 0.9683 |
| 0.0001 | 9.0 | 20394 | 0.1962 | 0.9722 |
| 0.0 | 10.0 | 22660 | 0.1638 | 0.9802 |
Framework versions
- Transformers 4.46.2
- Pytorch 2.5.1+cu121
- Datasets 3.1.0
- Tokenizers 0.20.3