BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//pretalx.com//euroscipy-2025//speaker//E3S7GV
BEGIN:VTIMEZONE
TZID:Europe/Warsaw
BEGIN:DAYLIGHT
DTSTART:20240818T000000
TZNAME:CEST
TZOFFSETFROM:+0200
TZOFFSETTO:+0200
END:DAYLIGHT
BEGIN:STANDARD
DTSTART:20241027T030000
RDATE:20251026T030000
TZNAME:CET
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20250330T030000
RDATE:20260329T030000
TZNAME:CEST
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
SUMMARY:Skrub: machine learning for dataframes - Guillaume Lemaitre\, Jér
 ôme Dockès\, Riccardo Cappuzzo
DTSTART;TZID=Europe/Warsaw:20250818T103000
DTEND;TZID=Europe/Warsaw:20250818T120000
DTSTAMP:20260911T202302Z
UID:pretalx-euroscipy-2025-8KK8UC@pretalx.com
DESCRIPTION:Machine-learning algorithms expect a numeric array with one ro
 w per observation. Typically\, creating this table requires "wrangling" wi
 th Pandas or Polars (aggregations\, selections\, joins\, ...)\, and to ext
 ract numeric features from structured data types such as datetimes. These 
 transformations must be applied consistently when making predictions for u
 nseen inputs\, and choices must be informed by performance measured on a v
 alidation dataset\, while preventing data leakage. This preprocessing is t
 he most difficult and time-consuming part of many data-science projects.\n
 \nSkrub bridges the gap between complex tabular data stored in Pandas or P
 olars dataframes\, and machine-learning algorithms implemented by scikit-l
 earn estimators. It provides scikit-learn transformers to extract features
  from datetimes\, (fuzzy) categories and text\, and to perform data-wrangl
 ing such as joins and aggregations in a learning pipeline. Its pre-built\,
  flexible learners offer very robust performance on many tabular datasets 
 without manual tweaking. It can create complex pipelines that handle multi
 ple tables\, while easily describing and searching rich hyperparameter spa
 ces. As interactivity and visualization are essential for preprocessing\, 
 Skrub also provides an interactive report to explore a dataframe\, and its
  pipelines can be built incrementally while inspecting intermediate result
 s.\n\nWe will give an overview of Skrub and demonstrate its features on re
 alistic and challenging tabular learning scenarios
LOCATION:Room 1.38 (Ground Floor)
URL:https://pretalx.com/euroscipy-2025/talk/8KK8UC/
END:VEVENT
END:VCALENDAR