This repository provides examples of surname classification using two different approaches: a Multilayer Perceptron (MLP) and a Convolutional Neural Network (CNN). Both models are trained to predict the nationality of a surname based on its characters.

1. Data Vectorization classes:
   • Vocabulary: Class to process text and extract vocabulary for mapping.
   • SurnameVectorizer: Vectorizes surnames and nationalities using the vocabularies.
2. Dataset class:
   • SurnameDataset: Loads the dataset, splits it into training, validation, and test sets, and generates batches.
3. Model:
   • SurnameClassifier: A 2-layer MLP for classifying surnames into nationalities.

1. Data Vectorization classes (similar to the MLP):
   • Vocabulary: Class to process text and extract vocabulary for mapping.
   • SurnameVectorizer: Vectorizes surnames and nationalities using the vocabularies.
2. Dataset class (similar to the MLP):
   • SurnameDataset: Loads the dataset, splits it into training, validation, and test sets, and generates batches.
3. Model:
   • SurnameCNNClassifier: A CNN model for surname classification.