<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
en2arCkptfromgendata_01
This model is a fine-tuned version of Botnoi/ckpt_marian_mt_en_ar_health on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.9214
- Bleu: 46.0887
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 |
---|---|---|---|---|
1.0453 | 1.0 | 151 | 0.9506 | 43.8903 |
0.8612 | 2.0 | 302 | 0.9097 | 45.4495 |
0.7031 | 3.0 | 453 | 0.8995 | 45.9366 |
0.6556 | 4.0 | 604 | 0.8995 | 46.2179 |
0.4798 | 5.0 | 755 | 0.9062 | 46.3055 |
0.4373 | 6.0 | 906 | 0.9074 | 46.6629 |
0.3865 | 7.0 | 1057 | 0.9136 | 45.9250 |
0.3516 | 8.0 | 1208 | 0.9175 | 46.1936 |
0.3819 | 9.0 | 1359 | 0.9205 | 46.0758 |
0.3803 | 10.0 | 1510 | 0.9214 | 46.0887 |
Framework versions
- Transformers 4.30.1
- Pytorch 2.0.1+cu118
- Datasets 2.12.0
- Tokenizers 0.13.3