marcusinthesky Goto Github PK

capitec-bblb-data-science-competition

This repository contains a finalist solution to the Capitec BBLB Data Science Competition hosted on Kaggle.

deepphenotypicmapping

Quantitative Imaging of a Mycobacterial Essential Gene Knockdown Library

dirtydata

DirtyData is blog about bad data and good Data Science. I don't know eveything but hopefully we will learnt together. 📓 Over the next few posts we will play with categorical data- the sparser the better- we will look at how to build data pipelines and cool techniques in exploring them. 🤓

duckface

A DuckDB Extention for the HuggingFace model repository.

ey-nextwave-data-science-challenge

This is a solution to the EY NextWave Data Science Challenge and placed as a South African National Finalist.

mlflow-helm

A template MLOPS infrastructure for model training and service.

peft

🤗 PEFT: State-of-the-art Parameter-Efficient Fine-Tuning.

postgresml

PostgresML is an end-to-end machine learning system. It enables you to train models and make online predictions using only SQL, without your data ever leaving your favorite database.

precarious-papers

An investigation in the impact of large public financial data breaches on financial markets and the role of Secret Offshore Special Purpose Vehicles in market pricing.

reciperobotapp

RecipeRobot is a app to inspire fun and novel food combinations. Choose an ingredient 🏪 and RecipeRobot will give you a recommendation on ingredients to pair with it using state-of-the-art methods in AI. RecipeRobot has learnt 🎓 from over 50 000 recipes the optimal pairings of ingredients to recommend the best and weirdest ones for you 🍉.

super-spirals

UCT Msc in Advanced Analytics and Decision Science project in modern Multivariate Analysis. This Project investigates the applications and failures of Variational Autoencoders.

talks

A collection of presentations given to various industry and research groups on topics of Machine Learning and Organizational Strategy.

tf-tabular

A flexible implementation of TabNet [https://arxiv.org/pdf/1908.07442.pdf] in Tensorflow 2.0.

word2risk

zindi-uber-ranrail-movement-challenge

Incident data in Cape Town, South Africa has been provided by SANRAL Freeway Management System and travel times between zones in Cape Town have been provided by Uber Movement. The aim of this challenge is to forecast if an incident will occur for each hour of each day per 500m road segment along the major roadways in Cape Town for 1 January 2019 to 31 March 2019.

zindi-umojahack-hackathon

A Zindi Hackathon.

zindi-umojahack-za

Zindi UmojaHack ZA was Data Science competition hosted for South Africa Universities in 25 July 2020. The competition looked to predict transit times for a ride hailing service.

zindi-urban-air-pollution

A solution to the the Zindi Africa Urban Air Pollution Challenge.

zindi-za-covid19-vulnerablity-map

An entry in the South African COVID-19 Vulnerability Map by #ZindiWeekendz.

zindi_loan_default

SuperLender is a local digital lending company, which prides itself in its effective use of credit risk models to deliver profitable and high-impact loan alternative. Its assessment approach is based on two main risk drivers of loan default prediction:. 1) willingness to pay and 2) ability to pay. Since not all customers pay back, the company invests in experienced data scientist to build robust models to effectively predict the odds of repayment