mt5-base-zu-en-mt is a machine translation model from isiZulu to English based on a fine-tuned google/mt5-base model. It establishes a strong baseline for automatically translating texts from isiZulu to English.
This model is limited by its training dataset. This may not generalize well for all use cases in different domains.
Specifically, this model is a mt5-base model that was fine-tuned on JW300 isiZulu corpus and LAFAND. The model was trained using isiXhosa (xh_ZA) as the language since the pre-trained model does not initially support isiZulu. Thus, you need to use the sw_KE for language code when evaluating the model.
This model was trained on NVIDIA V100 GPU
Fine-tuning mt5-base achieves 27.91 BLEU on LAFAND test set
By David Adelani
Select AutoNLP in the “Train” menu to fine-tune this model automatically.
- Downloads last month