opus-mt-de-en-OPUS_Medical_German_to_English

This model is a fine-tuned version of Helsinki-NLP/opus-mt-de-en.

Model description

For more information on how it was created, check out the following link: https://github.com/DunnBC22/NLP_Projects/blob/main/Machine%20Translation/Medical%20-%20German%20to%20English/OPUS_Medical_German_to_English_OPUS_Translation_Project.ipynb

Intended uses & limitations

This model is intended to demonstrate my ability to solve a complex problem using technology.

Training and evaluation data

Dataset Source: https://huggingface.co/datasets/ahazeemi/opus-medical-en-de

Histogram of German Input Word Counts

Histogram of English Input Word Counts

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
• gradient_accumulation_steps: 4
• total_train_batch_size: 32
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 5

Training results

• eval_loss: 0.8723
• eval_bleu: 53.88120
• eval_rouge:
  • rouge1: 0.7664
  • rouge2: 0.6284
  • rougeL: 0.7370
  • rougeLsum: 0.7370

• The training results values are rounded to the nearest ten-thousandth.

Framework versions

• Transformers 4.31.0
• Pytorch 2.0.1+cu118
• Datasets 2.14.4
• Tokenizers 0.13.3