BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//pretalx.com//euroscipy-2025//speaker//N7TBEV
BEGIN:VTIMEZONE
TZID:Europe/Warsaw
BEGIN:DAYLIGHT
DTSTART:20240820T000000
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:PyPI in the face: running jokes that PyPI download stats can play 
 on you - Loïc Estève
DTSTART;TZID=Europe/Warsaw:20250820T140500
DTEND;TZID=Europe/Warsaw:20250820T143500
DTSTAMP:20260915T030334Z
UID:pretalx-euroscipy-2025-LNU8UV@pretalx.com
DESCRIPTION:We all love to tell stories with data and we all love to liste
 n to them. Wouldn't it be great if we could also draw actionable insights 
 from these nice stories?\n\nAs scikit-learn maintainers\, we would love to
  use PyPI download stats and other proxy metrics (website analytics\, gith
 ub repository statistics\, etc ...) to help inform some of our decisions l
 ike:\n- how do we increase user awareness of best practices (please use Pi
 peline and cross-validation)?\n- how do we advertise our recent improvemen
 ts (use HistGradientBoosting rather than GradientBoosting\, TunedThreshold
 Classifier\, PCA and a few other models can run on GPU) ?\n- do users care
  more about new features from recent releases or consolidation of what alr
 eady exists?\n- how long should we support older versions of Python\, nump
 y or scipy ?\n\nIn this talk we will highlight a number of lessons learned
  while trying to understand the complex reality behind these seemingly sim
 ple metrics.\n\nTelling nice stories is not always hard\, trying to grasp 
 the reality behind these metrics is often tricky.
LOCATION:Room 1.20 (Ground Floor\, Shannon)
URL:https://pretalx.com/euroscipy-2025/talk/LNU8UV/
END:VEVENT
END:VCALENDAR