English to Tamazight (zgh) Translator Model

This model is a fine-tuned MarianMT model for translating text from English (en) to Tamazight (zgh). It is based on the Helsinki-NLP/opus-mt-en-ROMANCE architecture and fine-tuned using a custom dataset of English-Tamazight sentence pairs.

Model Details

Base Model: Helsinki-NLP/opus-mt-en-ROMANCE
Fine-Tuned Language Pair: English (en) → Tamazight (zgh)
Framework: Hugging Face Transformers
Architecture: MarianMT

Features

Input Language: English
Output Language: Tamazight (Standard Amazigh, ISO 639-3: zgh)
Use Case: Translation of text from English to Tamazight, suitable for low-resource machine translation tasks.

How to Use

Installation

Install the required libraries:

pip install transformers torch sentencepiece

Loading the Model

Use the following code to load and use the model for translation:

from transformers import MarianMTModel, MarianTokenizer

# Load the fine-tuned model and tokenizer
model_name = "ilyasaqit/MarianMTModel-EN-ZGH-PreTrained"
tokenizer = MarianTokenizer.from_pretrained(model_name)
model = MarianMTModel.from_pretrained(model_name)

# Translate text
def translate_to_tamazight(text: str) -> str:
    inputs = tokenizer(text, return_tensors="pt", padding=True, truncation=True)
    translated = model.generate(**inputs)
    return tokenizer.decode(translated[0], skip_special_tokens=True)

# Example usage
text = "Hello, how are you?"
tamazight_translation = translate_to_tamazight(text)
print(f"Tamazight Translation: {tamazight_translation}")