README.md

metadata

license: apache-2.0
tags:
  - generated_from_trainer
metrics:
  - bleu
model_index:
  - name: opus-mt-ja-en-finetuned-ja-to-en_test
    results:
      - task:
          name: Sequence-to-sequence Language Modeling
          type: text2text-generation
        metric:
          name: Bleu
          type: bleu
          value: 76.0826

opus-mt-ja-en-finetuned-ja-to-en_test

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

• Loss: 0.4001
• Bleu: 76.0826
• Gen Len: 16.4264

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: 0.0002
• 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

Training Loss	Epoch	Step	Validation Loss	Bleu	Gen Len
1.1048	1.0	247	0.5843	62.1912	16.2467
0.487	2.0	494	0.4414	71.5439	16.4904
0.2689	3.0	741	0.4001	76.0826	16.4264

Framework versions

• Transformers 4.9.1
• Pytorch 1.9.0+cu111
• Datasets 1.10.2
• Tokenizers 0.10.3