Instructions to use masakhane/m2m100_418M_fon_fr_rel_news with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use masakhane/m2m100_418M_fon_fr_rel_news with Transformers:
# Load model directly from transformers import AutoTokenizer, AutoModelForSeq2SeqLM tokenizer = AutoTokenizer.from_pretrained("masakhane/m2m100_418M_fon_fr_rel_news") model = AutoModelForSeq2SeqLM.from_pretrained("masakhane/m2m100_418M_fon_fr_rel_news", device_map="auto") - Notebooks
- Google Colab
- Kaggle
YAML Metadata Error:"datasets[0]" with value "JW300 + [LAFAND](https://github.com/masakhane-io/lafand-mt)" is not valid. If possible, use a dataset id from https://hf.co/datasets.
m2m100_418M-fon-fr-mt
Model description
m2m100_418M-fon-fr-mt is a machine translation model from Fon to French based on a fine-tuned facebook/m2m100_418M model. It establishes a baseline for automatically translating texts from Fon to French.
Limitations and bias
This model is limited by its training dataset. This may not generalize well for all use cases in different domains.
Training data
Specifically, this model is a m2m100_418M model that was fine-tuned on JW300 Fon corpus and LAFAND.
Training procedure
This model was trained on NVIDIA V100 GPU
Eval results on Test set (BLEU score)
Fine-tuning m2m100_418M achieves 7.81 BLEU on LAFAND test set
BibTeX entry and citation info
By David Adelani
- Downloads last month
- 6