README.md

metadata

license: apache-2.0
tags:
  - generated_from_trainer
datasets:
  - wmt16
metrics:
  - bleu
model-index:
  - name: opus-mt-tr-en-finetuned-en-to-tr
    results:
      - task:
          name: Sequence-to-sequence Language Modeling
          type: text2text-generation
        dataset:
          name: wmt16
          type: wmt16
          config: tr-en
          split: train
          args: tr-en
        metrics:
          - name: Bleu
            type: bleu
            value: 6.471

opus-mt-tr-en-finetuned-en-to-tr

This model is a fine-tuned version of Helsinki-NLP/opus-mt-tr-en on the wmt16 dataset. It achieves the following results on the evaluation set:

• Loss: 1.9429
• Bleu: 6.471
• Gen Len: 56.1688

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: 3
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Bleu	Gen Len
1.5266	1.0	12860	2.2526	4.5834	55.6563
1.2588	2.0	25720	2.0113	5.9203	56.3506
1.1878	3.0	38580	1.9429	6.471	56.1688

Framework versions

• Transformers 4.25.1
• Pytorch 1.13.1+cu116
• Datasets 2.8.0
• Tokenizers 0.13.2