Skip to content

We train a recurrent neural network on a dataset containing all the names of Pokémons and use the model to generate new names.

Notifications You must be signed in to change notification settings

yangobeil/Pokemon-name-generator

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

3 Commits
 
 
 
 

Repository files navigation

Pokemon-name-generator

We start by finding all the Pokémon names online on a website that has an API for Pokmon data. We then clean the data to ransform it into a form that the neural network can take.

The next step is to train a recurrent neural network with a LSTM layer on the data. This is then used to generate new names. A few nice examples of the results are:

  • oscalish
  • incono
  • unteri
  • pine

The model does surprisingly not to bad given the amount of data. Most of the names are coherent, way more than earlier in the training, and some of them would even be usable.

About

We train a recurrent neural network on a dataset containing all the names of Pokémons and use the model to generate new names.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published