README.md

metadata

license: apache-2.0
tags:
  - translation
  - generated_from_trainer
datasets:
  - kftt
metrics:
  - bleu
model-index:
  - name: marian-finetuned-kftt-ja-to-en
    results:
      - task:
          name: Sequence-to-sequence Language Modeling
          type: text2text-generation
        dataset:
          name: kftt
          type: kftt
          config: en-ja
          split: validation
          args: en-ja
        metrics:
          - name: Bleu
            type: bleu
            value: 19.353560365370512

marian-finetuned-kftt-ja-to-en

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

• Loss: 1.9124
• Bleu: 19.3536

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: 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: 3
• mixed_precision_training: Native AMP

Training results

Framework versions

• Transformers 4.27.4
• Pytorch 2.0.0+cu117
• Datasets 2.3.2
• Tokenizers 0.13.3