x-marian-finetuned-multidataset-kin-to-en

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

• Loss: 1.1731
• Bleu: 42.9597

Model Description

The model has been fine-tuned to perform machine translation from Kinyarwanda to English.

Intended Uses & Limitations

The primary intended use of this model is for research purposes.

Training and Evaluation Data

The model was fine-tuned using a combination of datasets from the following sources:

• Digital Umuganda
• Masakhane
• Muennighoff

For the training of the machine translation model, the dataset underwent the following preprocessing steps:

• Text was converted to lowercase
• Digits were removed

The combined dataset was divided into training and validation sets, with a split of 90% for training and 10% for validation.

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 2e-05
• train_batch_size: 32
• eval_batch_size: 64
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 3

Training results

Framework versions

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