Automated Deep Learning: Neural Architecture Search Is Not the End (a curated list of AutoDL resources and an in-depth analysis)

An autoML framework & toolkit for machine learning on graphs.

Towards Generalized and Efficient Blackbox Optimization System/Package (KDD 2021 & JMLR 2024)

DEEPScreen: Virtual Screening with Deep Convolutional Neural Networks Using Compound Images

A paper collection about automated graph learning

An efficient open-source AutoML system for automating machine learning lifecycle, including feature engineering, neural architecture search, and hyper-parameter tuning.

Students Performance Evaluation using Feature Engineering, Feature Extraction, Manipulation of Data, Data Analysis, Data Visualization and at lat applying Classification Algorithms from Machine Learning to Separate Students with different grades

Nature-inspired algorithms for hyper-parameter tuning of Scikit-Learn models.

Convenient classes for optimizing Hyper-parameters, using Random search, Spearmint and SigOpt

Combined hyper-parameter optimization and feature selection for machine learning models using micro genetic algorithms

A gradient free optimization routine which combines Particle Swarm Optimization with a local optimization for each particle

Grammaropt : a framework for optimizing over domain specific languages (DSLs)

Pipelineopt, sckit-learn automatic pipeline optimization

Hyper-Parameter Optimisation experiment as part of my undergraduate dissertation (2019)

Students Performance Evaluation using Feature Engineering, Feature Extraction, Manipulation of Data, Data Analysis, Data Visualization and at lat applying Classification Algorithms from Machine Learning to Separate Students with different grades

Pipoh is a library that implements several diversification techniques base on mean-variance framework. In addition, it includes a novel purely data-driven methods for determining the optimal value of the hyper-parameters associated with each investment strategy.

Python implementation that explores how different parameters impact a single hidden layer of a feed-forward neural network using gradient descent