MarianMT English to Indonesian (Fine-Tuned)

evaluation matrix

Model Description

This model is a fine-tuned version of the pre-trained Helsinki-NLP/opus-mt-en-id (MarianMT) model. It has been specifically trained to improve translation quality from English to Indonesian, focusing on [sebutkan fokus domain Anda di sini, misal: domain sains / percakapan sehari-hari / dact dataset].

  • Original Model: Helsinki-NLP/opus-mt-en-id
  • Architecture: MarianMT
  • Source Language: English (en)
  • Target Language: Indonesian (id)
  • License: Apache 2.0

Intended Uses & Limitations

This model is intended for direct text translation from English to Indonesian.

  • Best for: [ menerjemahkan teks sains, artikel formal, dll.]
  • Limitations: As with any machine translation model, it may struggle with highly idiomatic English phrases or highly informal Indonesian slang.

How to Use

You can easily use this model with the Hugging Face transformers library.

Using pipeline

from transformers import pipeline

# Load the fine-tuned model
translator = pipeline("translation", model="Ik45/marianMT-en-id-dact-finetuning")

text = "Quantum mechanics is a fundamental theory in physics."
result = translator(text)

print(result[0]['translation_text'])
Downloads last month
61
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Dataset used to train Ik45/mt-science-en-id-finetuning