metadata
license: apache-2.0
base_model: ihebaker10/spark-name-ar-to-en
tags:
- generated_from_trainer
metrics:
- bleu
- wer
model-index:
- name: spark-name-ar-to-en
results: []
spark-name-ar-to-en
This model is a fine-tuned version of ihebaker10/spark-name-ar-to-en on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.4228
- Bleu: 46.0889
- Wer: 0.513
- Cer: 0.3032
- Gen Len: 6.6541
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: linear
- num_epochs: 10
Training results
Training Loss | Epoch | Step | Validation Loss | Bleu | Wer | Cer | Gen Len |
---|---|---|---|---|---|---|---|
1.4673 | 1.0 | 2221 | 1.4345 | 47.0582 | 0.4936 | 0.2991 | 6.4236 |
1.3251 | 2.0 | 4442 | 1.4000 | 50.8792 | 0.477 | 0.2893 | 6.4104 |
1.0338 | 3.0 | 6663 | 1.3892 | 51.5038 | 0.4727 | 0.2872 | 6.418 |
0.9901 | 4.0 | 8884 | 1.3888 | 52.5143 | 0.4675 | 0.2821 | 6.3892 |
0.8191 | 5.0 | 11105 | 1.3953 | 52.9892 | 0.4664 | 0.2902 | 6.5155 |
0.805 | 6.0 | 13326 | 1.4024 | 52.8471 | 0.4654 | 0.2904 | 6.5205 |
0.7153 | 7.0 | 15547 | 1.4127 | 45.0993 | 0.5169 | 0.304 | 6.6268 |
0.673 | 8.0 | 17768 | 1.4138 | 45.2321 | 0.5186 | 0.3087 | 6.6599 |
0.6606 | 9.0 | 19989 | 1.4228 | 45.6946 | 0.5154 | 0.3059 | 6.6617 |
0.6626 | 10.0 | 22210 | 1.4228 | 46.0889 | 0.513 | 0.3032 | 6.6541 |
Framework versions
- Transformers 4.40.2
- Pytorch 2.2.1+cu121
- Datasets 2.19.1
- Tokenizers 0.19.1