Date / Heure
Date(s) - 03/10/2019
19h00 - 21h00
Emplacement
Scaleway
Catégories
Mark your calendar for the next session of the PyData Paris Meetup on October 3rd 2019. This Meetup will be hosted at Scaleway’s headquarters located 11bis Rue Roquépine, Paris.
The speakers for this session are Marianne Corvellec and Arnaud Wald.
Schedule
————-
7:00pm – 7:15pm: Community announcements
7:15pm – 8:00pm: Marianne Corvellec:
Building custom analytics web apps for bioinformatics with Plotly’s Dash Bio
8:00pm – 8:30pm: Arnaud Wald
RAPIDS.ai – Leveraging GPUs for accelerated data science & data analytics
8:30pm – 9:30pm: Standing buffet sponsored by our host, Scaleway
Bios
——
Marianne Corvellec
A PhD in statistical physics, Marianne works in industry as a data
scientist and a software developer. She is also a free software
activist (with April.org) and an independent researcher (with IGDORE).
Her research interests include data science, education, and
assessment. Since 2013, she has been a regular speaker and contributor
in the Python, Carpentries, and FLOSS communities.
Arnaud Wald
Arnaud Wald has been working as a Machine Learning Engineer since he graduated from Centrale-Supélec in 2017. Now a member of Scaleway’s AI team, he can keep up to date with the latest advances in the field while working on practical solutions to make AI in the cloud more accessible.
Abstracts
————-
Building custom analytics web apps for bioinformatics with Plotly’s Dash Bio
Dash Bio is a free and open-source library for creating customizable
data dashboards in the field of the life sciences (molecular biology,
analytical chemistry, …). These dashboards are interactive,
reactive, web-based applications, which can be written in either
Python or R. They let users perform bioinformatics-related tasks such
as exploring, analyzing, and visualizing genomic data.
First, we focus on the extensibility and Plotly compatibility of the library. Dash Bio is highly extensible, being one of several
domain-specific component libraries in the Dash ecosystem. Dash is a
Python-based web framework for building analytics (data science) web
applications. Running Flask in the backend and React.js in the
frontend, it leverages the power of Plotly graphs.
Dash has recently gained much interest and adoption in the scientific
and business communities. We then discuss how Dash Bio fits in the
bioinformatics space, from parsing utilities, at the lower level, to
research use cases, at the higher level. Finally, we demo how to write
a Dash app from scratch, in Python, using available Dash Bio
components and the very elegant Plotly Express.
RAPIDS.ai – Leveraging GPUs for accelerated data science & data analytics
RAPIDS makes it possible to have end-to-end data science pipelines run entirely on GPU architecture. It capitalizes on the parallelization capabilities of GPUs to accelerate data preprocessing pipelines, with a pandas-like dataframe syntax. GPU-optimized versions of scikit-learn algorithms are available, and RAPIDS also integrates with major deep learning frameworks. This talk will present RAPIDS and its capabilities, and how to integrate it in your pipelines.
Misc.
——-
We need all attendees to provide their full name as they register on meetup.com if it does not match their meetup username. Please bring a photo id with you to access the event.
https://www.meetup.com/fr-FR/PyData-Paris/events/263673958/