Simple, compact, and hackable post-hoc deep OOD detection for already trained tensorflow or pytorch image classifiers.

Critical difference diagrams with Python and Tikz

This repository is created for storing the components of Statistical Tests of One Pop, Two Pops and Three or more pops using Python.

Plot critical difference diagrams in Julia

code for "Spectral Removal of Guarded Attribute Information"

code for "Erasure of Unaligned Attributes from Neural Representations"

about statistical techniques for Data Science

Perform a STEP by STEP multiple mean comparison analysis on R

Parallel evolutionary algorithm that represents images with polygons using edge detection and denoising. Configuration is carried out through normality tests and pairwise comparisons.

ASIS is a web application developed for the compilation of impact report PDF documents for the tutorial programme (A-STEP) at the University of the Free State.

A set of statistical methods conducted on a strict set of algorithm's performance readings, utilizing Python

This project implements a Compact Letter Display (CLD) system in Python to simplify Tukey HSD results by assigning letters to groups

A collection of INF2178H course projects showcasing the full Data Science lifecycle, including experimental planning, data cleaning, exploratory analysis, modeling, and evaluation. Projects apply qualitative and quantitative methods for knowledge discovery and decision-making.

Analyze "winners" and "losers" of the Premier League's Project Restart through chi-square tests and post-hoc analysis to see which teams overperformed and underperformed.

An R project that investigates and visualizes the effect of sex and education on an individual's income level through the use of a full-factorial two-way ANOVA test and relevant post hoc significance tests conducted over a 2014 Pew Research Center dataset consisting of higher education attainment, gender, and income data.

This project analyzes a heart disease dataset to explore the relationship between cholesterol, heart rate, and chest pain type. It includes normality tests, outlier detection, correlation analysis, MANOVA, post-hoc tests, and VIF analysis, with visualizations using histograms, heatmaps, and boxplots.

Gain hands-on experience with ANOVA analysis, understanding its assumptions, and applying it to real-world datasets to understand differences among group means.

This project explores Tukey’s Honest Significant Difference test as a robust statistical method for comparing group means after conducting ANOVA. In real-world data analysis, we often need to determine not just whether groups are different, but which specific groups differ

This projects aim was to identify which crew members were most likely to survive the Titanic wreck based on features such as Work department, Gender, Age, etc. We use Machine Learning concepts such as Logistic Regression and feature selection as well as data preprocessing.