This is a pre-trained language translation model that aims to create a translation system for English and Swahili lanuages. It is a fine-tuned version of Helsinki-NLP/opus-mt-en-swc on an unknown dataset.
Model Details
Transformer architecture used
Trained on a 210000 corpus pairs
Pre-trained Helsinki-NLP/opus-mt-en-swc
2 models to enforce biderectional translation
Model Description
Developed by: Peter Rogendo, Frederick Kioko
Model type: Transformer
Language(s) (NLP): Transformer, Pandas, Numpy
License: Distributed under the MIT License
Finetuned from model [Helsinki-NLP/opus-mt-en-swc]: This pre-trained model was re-trained on a swahili-english sentence pairs that were collected across Kenya. Swahili is the national language and is among the top three of the most spoken language in Africa. The sentences that were used to train this model were 210000 in total.
This translation model is intended to be used in many cases, from language translators, screen assistants, to even in official cases such as translating legal documents.
Direct Use
pip install sentencepiece
from transformers import pipeline
model_checkpoint = "Rogendo/en-sw"
fine_tuned_model = pipeline("translation", model=model_checkpoint)
fine_tuned_model("Earlier today, I saw her going through the stalls in the market")
Use a pipeline as a high-level helper
from transformers import pipeline
pipe = pipeline("text2text-generation", model="Rogendo/sw-en")
Load model directly
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
tokenizer = AutoTokenizer.from_pretrained("Rogendo/sw-en")
model = AutoModelForSeq2SeqLM.from_pretrained("Rogendo/sw-en")
Downstream Use [optional]
[More Information Needed]
Out-of-Scope Use
[More Information Needed]
Bias, Risks, and Limitations
[More Information Needed]
Recommendations
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
How to Get Started with the Model
Use a pipeline as a high-level helper
from transformers import pipeline
pipe = pipeline("text2text-generation", model="Rogendo/sw-en")
Load model directly
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
tokenizer = AutoTokenizer.from_pretrained("Rogendo/sw-en")
model = AutoModelForSeq2SeqLM.from_pretrained("Rogendo/sw-en")
