Wolfram Language

Generate Pokémon Names Using a Neural Net

This example demonstrates how to generate new Pokémon names from existing names using a neural network.

Take a list of Pokémon names.

Take a trained language model from the neural net repository (here the English language model).

Perform net surgery for handling any sequence length and the ability to stop generating characters.

Train the net on the set of Pokémon names.

Take the part of the net that generates the next character given previous ones.

Define a function to efficiently generate a pseudo Pokémon name.

Generate new Pokémon.

Related Examples

de es fr ja pt-br zh