AI-102 · Question #471
AI-102 Question #471: Real Exam Question with Answer & Explanation
With only 900 parallel sentences, the dataset is below the 10,000 sentence minimum required for Full Training in Azure Custom Translator, making Dictionary-only training the appropriate choice since it uses the data as a glossary/dictionary to guide translations. The training dat
Question
Drag and Drop Question You have a custom model named Model1 in the Azure AI translator service. You have training data that contains 900 parallel phrases and sentences and industry-specific terminology. You need to use the training data to train Model1 to translate between English and French. You sign in to the Custom Translator portal. Which four actions should you perform in sequence from the portal? To answer, move the appropriate actions from the list of actions to the answer area and arrange them in the correct order. Answer:
Explanation
With only 900 parallel sentences, the dataset is below the 10,000 sentence minimum required for Full Training in Azure Custom Translator, making Dictionary-only training the appropriate choice since it uses the data as a glossary/dictionary to guide translations. The training data must be uploaded as a PDF file (or plain text/TMX/XLIFF formats supported by Custom Translator portal), and after training, the correct workflow is to test the model to evaluate quality, then publish it to make it available for use via the API. Dictionary-only training is specifically designed for smaller datasets that contain domain-specific terminology rather than sufficient parallel sentence pairs for full neural model training.
Topics
Community Discussion
No community discussion yet for this question.