README.org

Links to slides & materials from PyData Berlin 2018

Friday, 05.07.2018

Track 1

• 09:00-09:30 Using GANs to improve generalization in a semi-supervised setting - trying it in open datasets, Andreas Merentitis, Carmine Paolino, Vaibhav Singh
• 10:45-12:15 Deep Neural Networks with PyTorch, Stefan Otte
  • https://github.com/sotte/pytorch_tutorial
• 13:15-14:45 Scaling and reproducing deep learning on Kubernetes with Polyaxon, Mourad Mourafiq
• 15:00-16:30 Production ready Data-Science with Python and Luigi, Mark Keinhörster
  • https://github.com/crazzle/pydata_berlin_2018
• 16:45-18:15 Deploying a machine learning model to the cloud using AWS Lambda, Benjamin Weigel

Track 2

• 09:00-10:30 Tricks, tips and topics in Text, Analysis Bhargav Srinivasa Desikan
  • https://github.com/bhargavvader/personal/tree/master/notebooks
• 10:45-12:15 Leveling up your storytelling and visualization skills, Gerrit Gruben
  • https://github.com/uberwach/leveling-up-viz-story
• 13:15-14:45 A Hands-On Introduction to Your First Data Science Project Em Grasmeder, Jin Yang
  • https://github.com/ThoughtWorksInc/twde-datalab
• 15:00-16:30 Search Relevance: A/B testing to Reinforcement Learning, Arnab Dutta
  • https://github.com/kraktos/MAB
• 16:45-18:15 Deprecating the state machine: building conversational AI with the Rasa stack Justina Petraitytė

Saturday, 06.07.2018

Neural Networks

• 11:00-11:45 Visual concept learning from few images, Vaibhav Singh
• 11:45-12:30 Simple diagrams of convoluted neural networks, Piotr Migdał

NLP & Text Analysis

Python Applications

• 11:00-11:45 Five things I learned from turning research papers into industry prototypes, Ellen König
• 11:45-12:30 Spatial Data Analysis With Python, Dillon R. Gardner
  • https://github.com/dillongardner/PyDataSpatialAnalysis
• 12:30-13:15 Python Unittesting for Ethereum Smart Contracts or how not to create your own Cryptocurrency, Robert Meyer
  • https://github.com/SmokinCaterpillar/Snowballs

Tools of the Trade

• 11:45-12:30 Towards automating machine learning: benchmarking tools for hyperparameter tuning, Thorben Jensen
• 12:30-13:15 Launch Jupyter to the Cloud: an example of using Docker and Terraform, Cheuk Ting Ho

---

Algorithms

• 15:45-16:30 Solving very simple substitution ciphers algorithmically, Stephen Enright-Ward

NLP & RNNs

• 14:15-15:00 Building new NLP solutions with spaCy and Prodigy, Matthew Honnibal
• 15:00-15:45 How I Made My Computer Write it’s First Short Story, Alexander Hendorf
• 15:45-16:30 Understanding and Applying Self-Attention for NLP, Ivan Bilan

Python in the Field

• 14:15-15:00 Python in Medicine: analysing data from mechanical ventilators and patient monitors Gusztav Belteki
• 15:00-15:45 How to scare a fish (school), Andrej Warkentin

Unsupervised Learning & Visualization

• 14:15-15:00 Manifold Learning and Dimensionality Reduction for Data Visualization and Feature Engineering, Stefan Kühn
  • https://github.com/cc-skuehn/Manifold_Learning
• 15:00-15:45 Extracting relevant Metrics with Spectral Clustering, Evelyn Trautmann
  • https://github.com/metterlein/spectral_clustering/
• 15:45-16:30 On Laplacian Eigenmaps for Dimensionality Reductio, Juan Orduz

---

Lightning Talks

• The Five Tribes of Machine Learning Explainers, Michał Łopuszyński

Sunday, 07.08.2018

ML in Production

• 10.15-11:00 Industrial ML - Overview of the technologies available to build scalable machine learning, Alejandro Saucedo
  • https://github.com/axsauze/industrial-machine-learning
  • https://github.com/axsauze/crypto-ml

Explainability and Privacy

• 10:15-11:00 GDPR in practise - Developing models with transparency and privacy in mind, Łukasz Mokrzycki
• 11:00:11:45 Privacy-preserving Data Sharing, Omar Ali Fdal
• 11:45-12:30 pyGAM: balancing interpretability and predictive power using Generalized Additive Models in Python Dani Servén Marín

ML in Production

• 11:00-11:45 How mobile.de brings Data Science to Production for a Personalized Web Experience, Florian Wilhelm and Markus Schüler

Computer Vision & CNNs

• 10:15-11:00 When to go deep in Computer Vision… and how, Irina Vidal Migallón
• 11:45-12:30 The Face of Nanomaterials: Insightful Classification Using Deep Learning Angelo Ziletti

---

Best Practices

• 14:15-15:00 Data versioning in machine learning projects, Dmitry Petrov

Extending Python

• 13:30-14:15 Interfacing R and Python, Andrew Collier
• 14:15-15:00 Extending Pandas using Apache Arrow and Numba, Uwe L. Korn

Bayesian Methods

• 13:30-14:15 All that likelihood with PyMC3, Junpeng Lao
  • https://github.com/junpenglao/All-that-likelihood-with-PyMC3

---

Performance

• 15:15-16:00 Big Data Systems Performance: The Little Shop of Horrors, Jens Dittrich
• 16:00-16:45 Battle-hardened advice on efficient data loading for deep learning on videos, Valentin Haenel
  • https://github.com/TwentyBN/20bn-video-data-loading-talk

New Libraries

• 16:00-16:45 LightFields.jl: Fast 3D image reconstruction for VR applications Hector, Andrade Loarca

Visualization Tools

• 15:15-16:00 Meaningful histogramming with Physt, Jan Pipek
  • https://janpipek.github.io/pydata2018-berlin/
• 16:00-16:45 Practical examples of interactive visualizations in JupyterLab with Pixi.js and Jupyter Widgets, Jeremy Tuloup
  • https://github.com/jtpio/pixijs-jupyter
  • Binder version: https://mybinder.org/v2/gh/jtpio/pixijs-jupyter/pydata-berlin?urlpath=lab/tree/examples/presentation.ipynb

---

Lightning Talks

• Missing talk ;) Some tools to ease EDA, Stefan Otte

Notes, write-ups

• Notes by @liopic
  • https://gist.github.com/liopic/6f1a6d50d41bd07efc18c876329ab254
• Notes by jznf
  • https://gitlab.com/jznf/pydata-berlin-2018