/aqua

Random sentence generator trained on KonoSuba light novel using Markov chain.

Primary LanguageJupyter NotebookMIT LicenseMIT

Aqua

Random sentence generator trained on KonoSuba light novel using Markov chain.

Also included is a FastAPI webserver using Aqua to generate random sentences.

Training data is based on the one used by Kazuma.

Example

Normal sentences:

  • If that’s the case, I’ll wait for the Destroyer to pass through, and rebuild the town.
  • Kazuma, warn them too… Come to think of Kazuma as a terrorist, I’ll show you just how great I can be when I get serious!
  • This isn’t the capital of water and hot springs, Alcanretia?
  • This way, the gate to the mortal world… Keep this a secret, okay?
  • When the guild judges the quest to defeat the Demon King to me.

Short sentences:

  • Can you do that, I’ll share some mana with me!
  • I might look like a devil!
  • You defeated two Demon King’s army is here!
  • … Is… Is that so?
  • No matter how you look at it, isn’t this a rifle?

Usage

Clone the project.

$ git clone https://github.com/MarsRon/aqua

Create a Python virtual environment.

$ python3 -m venv venv
$ source venv/bin/activate

Install libraries.

$ pip install -r requirements.txt

Run the test script to generate a few sentences.

$ python test.py

Start the FastAPI webserver.

$ python main.py

Generate your own Markov chain

I've included a Jupyter notebook so that you can make your own Aqua.

See train.ipynb for more information.

License

Distributed under the MIT License. See LICENSE.md for more information.

Contact

MarsRon - marsron204@gmail.com - marsron.name.my