lightgbmのfeature-transform（特徴量の非線形化）をすることで、80,000を超える特徴量を線形回帰でも表現できることを示します

Tao Ju's original C++ implementation, with CMake (compiles on Linux)

An implementation of heightmap-based terrain generation algorithms.

Python implementation of the PAMI 2012 paper "Measuring the Objectness of Image Windows" and the CVPR 2010 paper "What is an object ?"

Author's implementation of SoftCon: Simulation and Control of Soft-Bodied Animals with Biomimetic Actuators (SIGGRAPH Asia 2019 Technical Paper)

Two-Stage Multithreshold Otsu method.

A greedy outlier algorithm based on entropy. Implementation of He Z., Deng S., Xu X., Huang J.Z. (2006) A Fast Greedy Algorithm for Outlier Mining.

WiscKey is a highly SSD optimized key-value storage based on LevelDB.

Software for the paper "dCSR: A Memory-Efficient Sparse Matrix Representation for Parallel Neural Network Inference"

This is part of the project Efficient identification of core and dead features in variability models

Minimum Degree Spanning Tree

ONNLab is a framework and project for developing obfuscated neural networks, offering full control over training and inference, flexible architecture configuration, and seamless integration of custom components. Its core principle is systematic yet highly flexible design, enabling implementation of any architecture or training strategy.