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
Practical DevOps for the busy data scientist
Search
Tania Allard
June 27, 2019
Technology
0
350
Practical DevOps for the busy data scientist
Tania Allard
June 27, 2019
Tweet
Share
More Decks by Tania Allard
See All by Tania Allard
The RSE hiring and career progression pipelines: Top tips to navigate them efficiently
trallard
0
140
Mentored Sprints - 2023
trallard
0
140
Mentored Sprints 2022 - kickoff
trallard
3
200
Como participar en el mercado emergente del codigo abierto
trallard
4
210
El presente y futuro del computo cientifico con Python
trallard
0
180
Foss for fun and profit
trallard
3
240
Open source for fun and profit: rethinking the long road of sustainability.
trallard
0
120
Docker and Python: making them play nicely and securely for Ml and DS
trallard
1
530
Python and Docker for ML and Data Science
trallard
1
370
Other Decks in Technology
See All in Technology
レガシーをぶっ壊せ。AEONで始めるDevRelの話 / Qiita Night 2024-2-22
aeonpeople
3
1.3k
GrafanaMeetup_AmazonManagedGrafanaのアクセス制御機能とマルチテナント環境下でのアクセス制御について
daitak
0
240
LangSmith入門―トレース/評価/プロンプト管理などを担うLLMアプリ開発プラットフォーム
os1ma
3
310
長期間TiDBを使ってきた話 @ 私たちはなぜNewSQLを使うのかTiDB選定5社が語る選定理由と活用LT / Experiences with TiDB Over Time
chibiegg
2
900
ゼロから始めるVue.jsコミュニティ貢献 / first-vuejs-community-contribution-link-and-motivation
lmi
1
130
require(ESM)とECMAScript仕様
uhyo
3
770
On Your Data を超えていく!
hirotomotaguchi
2
690
LayerXにおけるLLMプロダクト開発の今までとこれから
layerx
PRO
1
370
Azure Container Apps + Bicep 〜 こんな感じで運用しています
kaz29
2
480
FrontDoorとWebAppsを組み合わせた際のリダイレクト処理の注意点
kenichirokimura
1
530
ワールドカフェI /チューターを改良する / World Café I and Improving the Tutors
ks91
PRO
0
120
プロンプトエンジニアリングでがんばらない-Agentic Workflow へ-近藤憲児
kenjikondobai
3
850
Featured
See All Featured
How GitHub Uses GitHub to Build GitHub
holman
468
290k
Why Our Code Smells
bkeepers
PRO
331
56k
Building Adaptive Systems
keathley
31
1.9k
Refactoring Trust on Your Teams (GOTO; Chicago 2020)
rmw
25
2.3k
ParisWeb 2013: Learning to Love: Crash Course in Emotional UX Design
dotmariusz
104
6.6k
Visualization
eitanlees
136
14k
JavaScript: Past, Present, and Future - NDC Porto 2020
reverentgeek
40
4.4k
Music & Morning Musume
bryan
41
5.6k
Optimising Largest Contentful Paint
csswizardry
8
2.4k
Stop Working from a Prison Cell
hatefulcrawdad
266
19k
Automating Front-end Workflow
addyosmani
1356
200k
Put a Button on it: Removing Barriers to Going Fast.
kastner
58
3.1k
Transcript
Tania Allard, PhD @ixek Developer Advocate @Microsoft Practical DevOps for
the busy Data Scientist http://bit.ly/MancML-trallard
2 A bit of background never hurt anyone About us
3 @ixek
4 @ixek
5 Top top view… @ixek Stable model/application ready to be
productised R&D - develop, iterate fast, usually local or cloud Magic Is it live??
6 How I would like everything to work…. @ixek It
works…. now send it over to production R&D - develop, iterate fast, usually local or cloud Push code, tag, tag data* Worry free deployment! Wait and relax
7 @ixek
8
9 @ixek DevOps / DataOps / MLOps
10 DevOps is the union of people, process, and products
to enable continuous delivery of value into production What is DevOps anyway? @ixek
11 Sort of DevOps applied to data-intensive applications. Requires close
collaboration between engineers, data scientists, architects, data engineers and Ops. How does it fit for DS? @ixek
12 @ixek Aims to reduce the end-to-end cycle time of
data analytics/science from the origin of ideas to the creation of data artifacts.
13
14
15 7 steps to DS
16 Keep everything in source control - but allow for
experimentation
17
18 Standardize and define your environments in code (conda, pipfiles,
Docker)
19 Use canonical data sources - always know what data
you are using (where it comes and goes)
20
21 Automate wisely
22 https://xkcd.com/1205/
23
24 Use pipelines for repeatability and explainability
25 Deploy portable models
26
27 Test continuously and monitor production: shift left
28
29 Thank you @ixek http://bit.ly/MancML-trallard