You can deploy this model and start earning money today!

maanka2/
opus-mt-en-so-finetuned

Name: opus-mt-en-so-finetuned
Availability: InStock

maanka2/english-somali-parallel-corpus

Helsinki-NLP/opus-mt-synthetic-en-so

finetune

apache-2.0

Views

No views yet

OPUS-MT English to Somali Fine-Tuned

This model is a fine-tuned version of the English-to-Somali Seq2Seq translation model (opus-mt-en-so), trained using specialized parallel datasets. It has been optimized using bfloat16 precision and is designed to translate English text into natural, grammatical Somali.

Model Description

Developer: maanka2
Model Type: Seq2Seq Translation (Marian Architecture)
Language Pair: English to Somali (en-so)
License: apache-2.0
Primary Use Case: Text-to-Text generation for translation tasks

Training Details

The model was trained on parallel translation corpus data extracted from textual files aligned line-by-line.

How to Use

You can easily run inference with this model using the Hugging Face transformers library:

python

from transformers import AutoTokenizer, AutoModelForSeq2SeqLM

model_id = "maanka2/opus-mt-en-so-finetuned"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForSeq2SeqLM.from_pretrained(model_id)

text = "Hello, how are you today?"
inputs = tokenizer(text, return_tensors="pt", max_length=128, truncation=True)
outputs = model.generate(**inputs)
translation = tokenizer.decode(outputs[0], skip_special_tokens=True)
print(translation)