mcs_kfold stands for "monte carlo stratified k fold". This library attempts to achieve equal distribution of discrete/categorical variables in all folds. The greatest advantage of this method is that it can be applied to multi-dimensional targets.

Fast Online Triplet mining in Pytorch

Predicts which telecom customers are likely to churn with 95% accuracy using real-world data features from usage, billing, and support data. Implements Sturges-based binning, one-hot encoding, stratified 80/20 train-test split, and a two-level ensemble pipeline with soft voting. Achieves 94.60% accuracy, 0.8968 AUC, 0.8675 precision, 0.7423 recall.

使用比赛方提供的脱敏数据，进行客户信贷流失预测。

Three business analytics case studies were undertaken, encompassing market basket analysis, customer segmentation, and campaign management. SAS Visual Data Mining and Machine Learning on SAS Viya was utilized to explore data and provide insights. A comprehensive report addressing both technical and business aspects was delivered.

Python package for stratifying, sampling, and estimating model performance with fewer annotations.

An optimal stratified sample design for Commodity Flow Survey (CFS) based on Simulated Annealing and Genetic Algorithm. A script in Procedural PostgreSQL is used to generate a frame with 100,000 records based on publicly available data.

Data sampling library

Data consists of tweets scrapped using Twitter API. Objective is sentiment labelling using a lexicon approach, performing text pre-processing (such as language detection, tokenisation, normalisation, vectorisation), building pipelines for text classification models for sentiment analysis, followed by explainability of the final classifier

The objective is to analyze flight delays in the United States. Data from airlines, airports, and runways will be collected and processed. Machine learning models will be built using logistic regression, decision trees, and XGB classifiers. Visualizations will be created in Tableau, and Excel dashboards and SQL queries will be used for analysis.

This project focuses on applying advanced simulation methods for derivatives pricing. It includes Monte-Carlo, Variance Reduction Techniques, Distribution Sampling Methods, Euler Schemes, and Milstein Schemes.

Stratification of multi-label datasets

Data sampling library

Data sampling library

WiDS Datathon 2020 on patient health through data from MIT’s GOSSIS (Global Open Source Severity of Illness Score) initiative.

A C library with Python bindings for efficient stratified random sampling from binary buffers or files.

Perform Data Sampling with Python

Morph & Split is a web app designed for augmenting images and masks, performing stratified dataset splitting, and preprocessing image-mask pairs — all optimized for efficient machine learning and computer vision workflows.

Web scraper to get professor information, and a mass emailer that sends a website with a survey.