PyData Eindhoven 2020
2020-10-07..2020-10-09
-- online --
[PyData Eindhoven 2019, PyData Eindhoven 2020, PyData Eindhoven 2021, PyData Eindhoven 2022, PyData Eindhoven 2023, PyData Eindhoven 2024, PyData Eindhoven 2025]
who it's for
data scientists, machine learning engineers, data engineers, MLOps engineers, research scientists, python developers, software engineers
topics
RAPIDS GPU dataframes and cuML; Dask at global scale; managed Dask (Coiled); real-time ML with Kafka and Kubeflow; serverless/FaaS DevOps and TOSCA templates; probabilistic programming with Stan; causal inference with pymc3 and instrumental variables; graph machine learning and graph convolutions; notebook to production (Flask API + React); hypergeometric significance networks (HNet); accelerated MRI with deep learning; docker image size reduction for data science
| title | author | length |
|---|---|---|
| Improving Data Understanding with Graphical Hypergeometric Networks | dr.ir. E.Taskesen | 34m |
The talk introduces graphical hypergeometric networks (HNet), a method that tests associations between variables for statistical significance using… | ||
| RAPIDS: How to accelerate your data science pipeline by orders of magnitude | Adam Grzywaczewski | 34m |
RAPIDS is a suite that runs end-to-end data science on GPUs using cuDF, cuML, cuGraph and Dask with Apache Arrow and UCX plumbing, offering drop-in replacement… | ||
| Productivity and performance in one programming language - A tour of Julia | Avik Sengupta | 30m |
Claim: Julia eliminates the two‑language problem by letting the same language express low‑level kernels and high‑level models, using multiple dispatch and… | ||
| Interview about Codam Coding College | Corinne Vigreux | 29m |
Founding story of a navigation product that sold from zero to one million and grew to a €1.8 billion business, plus why a free coding school adopts the 42… | ||
| Building a coronaradar | Damiaan Zwietering | 01h 05m |
A working coronaradar that pulls EU CDC spreadsheets into pandas and reproduces country curves, double-log linearization of cumulative cases to fit a… | ||
| RADON: DevOps framework to optimally exploit serverless computing tech... | Damian Andrew Tamburri | 35m |
RADON presents an EU funded 30‑month DevOps framework that claims to help teams adopt serverless/FaaS while avoiding vendor lock‑in and lowering cost by… | ||
| Stan: why does it exist? when is it useful? why do I use it? | Daniel Lee | 28m |
Stan exists as a statically typed modeling language plus fast inference and interfaces (C++ autodiff core, HMC sampling, ADVI) to let users encode any… | ||
| Monitoring a TV streaming service with AI - from PageRank to graph convolutions | Dennis Ramondt | 33m |
Liberty Global monitors a 4K streaming platform serving over a million customers using a telemetry streaming stack, Spark to scale Facebook Prophet for anomaly… | ||
| Real-Time ML services: From ideation to production | Ilias Sarantopoulos | 30m |
Forecasting time series for a Real Time Dynamic Pricing service with 5–10 minute aggregated windows, a batch training job plus a streaming job consuming from… | ||
| From notebook to production using | Koen Vossen | 01h 00m |
Code converts a notebook into a Flask API, a React frontend and a live Heroku deployment, with the notebook refactored into domain models, repository… | ||
| Dask at Global Scale with | Matthew Rocklin | 31m |
Claim: Coiled offers managed Dask so users get a real cluster from two commands — pip install coiled and a couple Python calls — with software environments… | ||
| 10x smaller docker containers for Data Science | Matthijs Brouns | 32m |
Starts from a 1.7 GB baseline and actually reduces a PyData image with Matplotlib Scipy Numpy and Scikit-learn down to ~188 MB using concrete steps: use… | ||
| Artificial intelligence to accelerate MRI scans - Philips & LUMC winners in the... | Nicola Pezzotti | 34m |
Philips & LUMC win the fastMRI challenge by submitting an adaptive compressed sensing network that fuses deep learning with Fourier/physics-based steps, uses… | ||
| Productionizing an unsupervised machine learning model to... | Nikki van Ommeren, Maike Fischer | 30m |
Claims a usable pipeline that clusters open-text feedback so analysts avoid reading ~5,000 items a month, built with open-source Python and deployed in Docker… | ||
| Applying Deep Learning for Object Detection from a Telecom Perspective | Wilder Rodrigues | 32m |
VodafoneZiggo builds an image pipeline that collects 47,000 images, prunes ~10k with OpenCV variance of Laplace and HOG cosine checks, labels ~8,000 then… | ||
| Understanding Deep Neural Networks | Vladimir Osin | 27m |
Clear claims and concrete tools: model attributions use captum on a pretrained ResNet50 with top‑3 outputs, showing gradients (saliency, input×gradient, guided… | ||
| Playing by the Rules-Based-Systems | Vincent D Warmerdam | 38m |
Rule-based patterns in spacey produce labeled examples that train a machine learning model and reveal concrete differences: the ML model shows higher precision… | ||
| Bias in Data | Sundas Khalid | 25m |
Data claims include concrete numbers and failures: 3.7 billion users generate 40 trillion gigabytes yet under 1% is analyzed, Facebook job ads target 85% of… | ||
| The causal problem of overexposure to repetitive ads: instrumental variables and pymc3 | Ruben Mak | 17m |
Analyzes why targeted online advertising can lead to users being overexposed to repetitive ads by framing it as a causal inference problem. Uses a case study… | ||
| The industrial challenge of missing data | Reza Sahraeian | 26m |
Overview covers origins of missingness and concrete Python implementations, showing scikit‑learn throws errors on NA while LightGBM and XGBoost handle… | ||
| Fighting COVID with Python | Daan de Bruin | 34m |
Claims a national standardized ICU database of 15 million measurements aggregated from five hospitals, built with a PostgreSQL staging DB and reproducible… | ||
| nothing matches | ||