metadata

license: apache-2.0
tags:
  - generated_from_trainer
datasets:
  - un_multi
metrics:
  - bleu
model-index:
  - name: >-
      opus-mt-en-ar-evaluated-en-to-ar-4000instances-un_multi-leaningRate2e-05-batchSize8-11-action-1
    results:
      - task:
          name: Sequence-to-sequence Language Modeling
          type: text2text-generation
        dataset:
          name: un_multi
          type: un_multi
          args: ar-en
        metrics:
          - name: Bleu
            type: bleu
            value: 51.7715

opus-mt-en-ar-evaluated-en-to-ar-4000instances-un_multi-leaningRate2e-05-batchSize8-11-action-1

This model is a fine-tuned version of Helsinki-NLP/opus-mt-en-ar on the un_multi dataset. It achieves the following results on the evaluation set:

Loss: 0.1850
Bleu: 51.7715
Meteor: 0.5164
Gen Len: 25.5612

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: 8
eval_batch_size: 8
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 11

Training results

Training Loss	Epoch	Step	Validation Loss	Bleu	Meteor	Gen Len
0.6999	0.25	100	0.1959	50.1492	0.508	25.2788
0.1994	0.5	200	0.1931	51.003	0.513	25.4038
0.1863	0.75	300	0.1864	51.3268	0.5145	25.1675
0.1826	1.0	400	0.1841	51.2507	0.513	25.2388
0.1494	1.25	500	0.1840	51.4291	0.5159	25.4225
0.1483	1.5	600	0.1839	51.2645	0.5126	25.395
0.1547	1.75	700	0.1837	51.7589	0.5157	25.48
0.1487	2.0	800	0.1845	51.896	0.5177	25.3988
0.1235	2.25	900	0.1852	52.0583	0.5177	25.5212
0.1164	2.5	1000	0.1850	51.7715	0.5164	25.5612

Framework versions

Transformers 4.18.0
Pytorch 1.11.0
Datasets 2.1.0
Tokenizers 0.12.1