Good quality network connectivity is ever more important. For hybrid fiber coaxial (HFC) networks, searching for upstream \emph{high noise} in the past was cumbersome and time-consuming. Even with machine learning due to the heterogeneity of the network and its topological structure, the task remains challenging. We present the automation of a simple business rule (largest change of a specific value) and compare its performance with state-of-the-art machine-learning methods and conclude that the precision@1 can be improved by 2.3 times. As it is best when a fault does not occur in the first place, we secondly evaluate multiple approaches to forecast network faults, which would allow performing predictive maintenance on the network.
May 10, 2022
Mar 14, 2022
State-of-the-art Deep Learning library for Time Series and Sequences in Pytorch / fastai. Recently I started to contribute to this project.
Sep 12, 2021
Easy configuration handling for complex machine learning pipelines
May 8, 2021
Free deep-learning in the cloud
Feb 21, 2018
https://docs.google.com/presentation/d/1VgocCul4ZyLVDqqr6iXsBRmmwsGhJkZj6-lzgdXs0ag/edit?usp=sharing A similar talk was held at a Meetup in December 2018 in Vienna using the following slides and focusing on sparklyR: https://docs.google.com/presentation/d/1NHG7-WoEUsjrdxFjy01OmZjxWB-FZomhfrxO-QapzKg/edit?usp=sharing There is even a videom from the talk. It was one of my first video recorded talks ;)
Sep 27, 2017
Dec 10, 2016
Individual cost based classification model outperforms classical processes.
Apr 27, 2016
Reverse engineering old data pipelines ; ) and analyzing huge quantities of spatial data.
Apr 27, 2016