This project utilizes Generative Adversarial Networks (GANs) to tackle the problem of credit card fraud detection. GANs are a powerful deep learning technique that can be used for generating synthetic data, which can be beneficial in situations with imbalanced datasets, such as fraud detection.

A tool to generate synthetic dataset of corporate travels

JR (jrnd.io) Source Connector for Apache Kafka Connect

An add-on for Blender that allows you to generate and render three-dimensional scenes that can be automatically annotated and used for training neural networks.

Generates structured synthetic datasets using Pydantic v2 for validation and Azure OpenAI for generating realistic open string data. It supports configurable data schemas, distributions, dependencies, and prompts.

[DEMO] "Pose Estimation in Gebäuden anhand von Convolutional Neural Networks und simulierten 3D-Daten" as an example of PDF-generation via LaTeX - good old days 2019, supervised by Prof. Dr.-Ing. Markus Könug & M. Sc. Patrick Herbers | click here

Fraud detection model

Implementation of BISECT from [2] for Hidden Outlier generation.

SensibleSleep is an open-source Python package that implements a Hierarchical Bayesian model for learning sleep patterns from smartphone screen-on events.

Synthetic Credit Card Transaction Generator used in the Sparkov program.

Code to generate skin cancer syntetic images