Skip to content

A Simple and intuitive deep learning library with high-efficiency, modularization and expandability.

Notifications You must be signed in to change notification settings

CelestialPaler/DeepLearningDevelopingKit

Repository files navigation

Deep Learning Developing Kit

  An algorithm library for deep learning.

Build status

Documentation Windows CPU Windows GPU
Documentation Build Status Build Status

Brief Introduction

    Powerful modularization and expandability provide a great tool to learn about machine learning, especially deep learning.You can explore as you wish, build your dream software, without being bothered by intricate codes. You can build your own AI model quickly and train it through iterator in high-efficiency.Like Lego building blocks, you can assemble or transplant your models and apply them to your AI applications within several minutes.

  Programing Language:

  • C/C++

  • Python

  Feature:

  • Simple and intuitive

  • High-efficiency

  • Modularization

  • Expandability

  • Customization

  Data-interchange Format:

  • Json

Algorithms

  • Regresstion Analysis

  • Decision Tree

  • Neural Network

  • Supported Vector Machine (Todo)

  • Expectation Maximization (Todo)

  • Clustering (Todo)

  • Decision Tree

  This project is mainly focusing on Deeplearning for now.

Developing Status

  Under developing:

  • Convolutional Neural Network

  • Refine Backpropagation Neural Network

  • MultiThread for acceleration

  • OpenCV support for image processing

  Further plan (Not top priority):

  • Visualization tool using SFML or Python

  • Acceleration using CUDA

About this project

    This is still an early stage of development. So, it`s welcomed for anyone to contribute to this project. Feel free to upload your code or pull a request. More than anything, please share your idea and give me some precious advices. XD

    For more information, check my website. Celestial Tech

    Copyright © 2015-2018 Celestial Tech Inc.

About

A Simple and intuitive deep learning library with high-efficiency, modularization and expandability.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published