https://github.com/akikurisu/n-gram-unity

N-Gram implementation in Unity using JobSystem
https://github.com/akikurisu/n-gram-unity

Last synced: 16 days ago
JSON representation

N-Gram implementation in Unity using JobSystem

• Host: GitHub
• URL: https://github.com/akikurisu/n-gram-unity
• Owner: AkiKurisu
• License: mit
• Created: 2023-09-08T06:41:25.000Z (over 1 year ago)
• Default Branch: main
• Last Pushed: 2023-11-07T13:21:06.000Z (about 1 year ago)
• Last Synced: 2023-11-08T04:21:16.504Z (about 1 year ago)
• Language: C#
• Size: 25.4 KB
• Stars: 1
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# N-Gram

This is an implementation in Unity for N-Gram prediction.

Article can be found in

[Implementing N-Grams for Player Prediction, Procedural Generation, and Stylized AI](http://www.gameaipro.com/GameAIPro/GameAIPro_Chapter48_Implementing_N-Grams_for_Player_Prediction_Proceedural_Generation_and_Stylized_AI.pdf) by Joseph Vasquez II

Using Job System to support muti-thread inference.

## Algorithm

A pseudocode algorithm for finding which N value provides the highest

prediction accuracy for pre-existing event sequences.

```

Instantiate an N-gram for each candidate N value ranged [1,20]

Create accuracy result storage for each N-gram

For each event sequence,

    For each event e ∈[0,L-1] in the sequence,

        For each N-gram ng,

            store result of e compared to ng.Predict()

            ng.UpdateWithNewEvent(e)

output accuracy results

output N value of most accurate N-gram based on results

```