metadata
license: apache-2.0
tags:
- generated_from_trainer
datasets:
- opus_infopankki
metrics:
- bleu
model-index:
- name: opus-mt-ar-en-finetuned-ar-to-en
results:
- task:
name: Sequence-to-sequence Language Modeling
type: text2text-generation
dataset:
name: opus_infopankki
type: opus_infopankki
args: ar-en
metrics:
- name: Bleu
type: bleu
value: 44.5107
opus-mt-ar-en-finetuned-ar-to-en
This model is a fine-tuned version of Helsinki-NLP/opus-mt-ar-en on the opus_infopankki dataset. It achieves the following results on the evaluation set:
- Loss: 0.9956
- Bleu: 44.5107
- Gen Len: 14.6465
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-06
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 11
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Bleu | Gen Len |
|---|---|---|---|---|---|
| 1.696 | 1.0 | 950 | 1.4125 | 35.5899 | 15.0109 |
| 1.4952 | 2.0 | 1900 | 1.2969 | 38.1745 | 14.7736 |
| 1.3821 | 3.0 | 2850 | 1.2175 | 39.8192 | 14.7649 |
| 1.2941 | 4.0 | 3800 | 1.1571 | 40.9966 | 14.7446 |
| 1.2409 | 5.0 | 4750 | 1.1097 | 41.9215 | 14.7406 |
| 1.1938 | 6.0 | 5700 | 1.0721 | 42.868 | 14.6819 |
| 1.1634 | 7.0 | 6650 | 1.0440 | 43.4749 | 14.6536 |
| 1.1355 | 8.0 | 7600 | 1.0223 | 43.9275 | 14.6809 |
| 1.1161 | 9.0 | 8550 | 1.0075 | 44.2378 | 14.6471 |
| 1.1103 | 10.0 | 9500 | 0.9987 | 44.4293 | 14.639 |
| 1.1043 | 11.0 | 10450 | 0.9956 | 44.5107 | 14.6465 |
Framework versions
- Transformers 4.19.2
- Pytorch 1.7.1+cu110
- Datasets 2.2.2
- Tokenizers 0.12.1