Upgrade to Pro
— share decks privately, control downloads, hide ads and more …
Speaker Deck
Features
Speaker Deck
PRO
Sign in
Sign up for free
Search
Search
Snotelier
Search
Tom Baldwin
February 15, 2016
0
170
Snotelier
A system for climbers and backcountry skiiers to preview avalanche danger from home.
Tom Baldwin
February 15, 2016
Tweet
Share
Featured
See All Featured
No one is an island. Learnings from fostering a developers community.
thoeni
21
3.5k
How To Stay Up To Date on Web Technology
chriscoyier
791
250k
Save Time (by Creating Custom Rails Generators)
garrettdimon
PRO
32
1.6k
個人開発の失敗を避けるイケてる考え方 / tips for indie hackers
panda_program
114
20k
Making Projects Easy
brettharned
119
6.4k
Visualizing Your Data: Incorporating Mongo into Loggly Infrastructure
mongodb
48
9.7k
StorybookのUI Testing Handbookを読んだ
zakiyama
31
6.2k
jQuery: Nuts, Bolts and Bling
dougneiner
64
7.9k
YesSQL, Process and Tooling at Scale
rocio
173
14k
Improving Core Web Vitals using Speculation Rules API
sergeychernyshev
19
1.2k
The Success of Rails: Ensuring Growth for the Next 100 Years
eileencodes
46
7.6k
Scaling GitHub
holman
463
140k
Transcript
snotelier automated avalanche danger summaries Mark Gallup Tom Baldwin Insight
Data Science Willamette Pass Ski Patrol
None
? ?
measure snowpack properties once per hour 75 sites in WA,
81 in OR ? ? SNOTEL
measure snowpack properties once per hour 75 sites in WA,
81 in OR ? ? SNOTEL
measure snowpack properties once per hour 75 sites in WA,
81 in OR ? ? SNOTEL ? ?
measure snowpack properties once per hour 75 sites in WA,
81 in OR ? ? SNOTEL ? ? snotelier automatic avalanche danger summaries from SNOTEL data
feature engineering
Feature Value 1-day snowfall 0.0 3-day snowfall 8.0 max rate
of snowfall 4.25 1-day water equivalent -0.1 3-day water equivalent 0.8 max rate of water equiv. 0.3 temperature during storm 31.0 temp. trend during storm 0.0 feature engineering condense time series data to set of relevant features
labeling + validation human-generated zone advisories from Northwest Avalanche Center
fit with logistic regression model
algorithm
algorithm
who benefits? • Vastly expanded coverage area relative to NWAC
(now covers OR) • Generated summary easier to digest • Tool for existing avalanche forecasters to prioritize areas
Tom Baldwin http://snotelier.com • Physics PhD 2015, University of Oregon
• Semiconductor optics in high magnetic fields • Running, climbing, snowboarding baldwint.com