Erdogan Taskesen bio photo

Erdogan Taskesen

Data Scientist in Statistical Machine Learning

Email Google Scholar LinkedIn Github Stackoverflow Orcid Medium Gumroad

Software

Overview

A small selection of my open-source projects that can be found on my github page:

d3blocks D3Blocks builts on the graphics of D3 Javascript to create the most visually attractive and useful charts with only a few lines of Python code.

distfit is for probability density function fitting of univariate distributions of non-censored data.

findpeaks is Python package for the detection of peaks and valleys in 1 dimensional vectors and (2D) images.

bnlearn is for learning the graphical structure of Bayesian networks in Python.

hgboost is to minimize the function for xgboost, catboost or lightboost over a hyper-parameter space by using cross-validation, and evaluating the results on an independent validation set. hgboost can be applied for classification and regression tasks.

classeval is to evaluate the models performance for any kind of two-class or multi-class model.

hnet stands for graphical Hypergeometric NEtworks, which is a method where associations across variables are tested for significance by statistical inference.