Johannes Horn's Projects
datamining, linear regression, multinomial logistic regression, cnn´s, gan´s, rnn´s, RL etc.
Predicting california_houseprices, using different ML/deeplearning methods.
command cheatsheet
This repository implements multiple Deep-RL agents to solve an 'openai-gym' puzzle. The idea is to share agent learned parameters to solve the puzzle ASAP!
tensorflows deeplab trained on a custom dataset
Endslate.AI is a software project which deals with workflow simplification in the film editing process. It automatically examines film and audio material and synchronizes them based on the clapperboard.
Enhances-Super-Resolution-Generative-Adversarial-Networks modified to run in google colab
A simple Android App which helps to name fotos the way you want.
Fusee aims at becoming a multiplatform 3D realtime engine with a strong emphasis on content transformation and manipulation. Fusee is written mainly in C# and can be programmed in C#. Fusee builds bridges to make your code run in environments that don't support C#, like Content-Creation-Software, C++, Plug-In-APIs and native HTML5/WebGL Javascript code.
ion flutter app
Predict the current energy consumption of a single household in London from the "Smart meters London" Dataset.
Implementation for "Deeper Task-Specificity Improves Joint Entity and Relation Extraction"
A simple Android app which is able to count your push-ups, gives you some execise suggestions and saves your progress. It uses posenet, a pre-trained deep learning model, to estimate body poses in real time. The key points detected by posenet are used to count your push-ups.
Visualizing some race data
The RWTH extensible training framework for universal recurrent neural networks
Aims to analyze if the method called "Sample Injection for Reinforcement Learning", in general can be helpful for different reinforcement learning algorithms and evironments to achieve better and/or faster results.
basic scheme interpreter from scratch written in C
Let a spider-like-robot teach itself to walk.
A masks to diameter service based on deeplabV3+ to calculate the diameter of a trunk from the output of deeplabV3+, trained with treeo dataset.