marian-finetuned-Umuganda-Dataset-en-to-kin-Umuganda-Dataset

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

• Loss: 1.8769
• Bleu: 32.8345

Model Description

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

Intended Uses & Limitations

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

Training and Evaluation Data

The model has been fine-tuned using the Digital Umuganda dataset.

The dataset was split with 90% used for training and 10% for testing.

The data used to train the model were cased and digits removed.

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