DataPredict

Welcome to Aqwam’s DataPredict Library!

By using or possessing any copies of this library or its assets (including the icons), you agree to our Terms and Conditions. In short:

• Attribution is required when using this library.

• A separate agreement is needed for commercial use (both internally and externally). This is applicable to companies (or individuals if none) with a revenue of at least $1000 in a lifetime, or a player base of at least total of 100 active players in a game from all servers, or those engaged in business-to-business activities.

• If you or your company is applicable to the above statement and do not want a separate agreement, then you are required to follow the “Commercial Use Conditions” in Terms And Conditions.

• Plus some more stuff…

Version	Current Version Number
Release	1.21
Beta	1.36.0

DataPredict is a machine and deep learning library for Roblox and Pure Lua.

This project is created to address the lack of machine and deep learning libraries as well as the inflexibility of the available ones on Roblox. It is also a way for me to apply my machine and deep learning knowledge into its physical form.

If there are any suggestions or issues for this library, don’t be afraid to reach out to me at my Discord server, DevForum thread or LinkedIn.

In addition, if you are interested in Roblox tutorials and future projects by me, then you can subscribe to me in YouTube.

To know what this library is all about from scratch, let’s have a look at the quick introduction here.

Tutorials

Installation

Installation For Roblox

The Basics - Level 1

Creating Our First Model

Saving And Loading Model Parameters

The Basics - Level 2

Using Optimizers

Using Regularizers

Using Gradient Descent Modifiers

Using Model Checker

Distributed Training

The Neural Networks

Using Neural Networks Part 1

Using Neural Networks Part 2

General Guide For Neural Networks Settings

Learning AIs - Level 1

Getting Started With Reinforcement Learning

Stimulating Curiosity With Random Network Distillation

Choosing The Most Effective Environment Inputs

Learning AIs - Level 2

Deep Reinforcement Learning Algorithms Properties

The Action Space And The Reinforcement Learning Algorithms

Example Codes

Learning AIs

Simple Environment

Self-Learning Sword-Fighting AIs - Version 8

The API

API Reference

API Reference

Want To Create Your Own Models, Optimizers And More?

API Design

Links

Download Links

Repository

DataPredict Library

MatrixL Library

Other Documentations

Aqwam’s Matrix Library

Version History

Release

Beta

All

Fun Facts About The DataPredict’s Icon

The three blocks in the icons represents these meanings, which also show’s the DataPredict’s Library main design principles:

• Basic: The icon uses simple blue blocks. This represents that the components of this library are very basic to use.

• Stackable: The icon shows the blocks are stacking on each other. This represents the components in this library can be stacked over each other.

• Portable: The three blocks can be unstacked and can move the individual blocks somewhere else. This represents how portable each of the components are when transfering from one system to another.

Blue colour is chosen as it represents knowledge. This represents the data.

White octagon makes us focus on the blue blocks. This represent the focus on the components of this library.