Upgrade to Pro
— share decks privately, control downloads, hide ads and more …
Speaker Deck
Speaker Deck
PRO
Sign in
Sign up
for free
Learning Data Science
Amit Kapoor
February 25, 2017
Science
0
430
Learning Data Science
A thoughtful approach for building your Data Science skills and portfolio.
Amit Kapoor
February 25, 2017
Tweet
Share
More Decks by Amit Kapoor
See All by Amit Kapoor
amitkaps
0
7
amitkaps
1
91
amitkaps
0
440
amitkaps
0
540
amitkaps
1
230
amitkaps
1
110
amitkaps
2
250
amitkaps
0
160
amitkaps
2
370
Other Decks in Science
See All in Science
ymdpharm
0
460
shuntaros
0
150
takehikoihayashi
0
1.4k
yuya4
3
380
event2020
0
150
shuntaros
0
500
asei
0
400
shuntaros
0
160
ianozsvald
0
280
kamakiri1225
0
660
kamakiri1225
0
610
lcolladotor
0
750
Featured
See All Featured
davidbonilla
70
3.5k
ufuk
56
5.4k
deanohume
295
27k
colly
187
14k
roundedbygravity
241
21k
jlugia
216
16k
dougneiner
55
5.4k
afnizarnur
176
14k
caitiem20
308
17k
jcasabona
7
520
rmw
11
740
revolveconf
200
9.6k
Transcript
Learning Data Science A thoughtful approach for building your Data
Science Skills Amit & Bargava
Data Science as a Craft
Art - Craft - Science
craft | krɑːft noun: An activity involving skill in making
or doing things by hand
How do you learn a craft?
Seek Inspiration
Start with Basics
Do Repetitions
Conduct Deliberate Practice
Focus on Mastery
Learning Data Science Ten guides for skill building
Shailii Understand your learning style. Choose the right approach -
Hacker v/s Structured. Online v/s Classroom. Self-driven v/s Mentor-driven.
Sangha Find your own community. Most of our learning happens
through peers. Find a set of data science peers that you can engage with.
Gunas Understand the basics. Start with learning the math &
statistics needed. Build an intuition on how the algorithms & the libraries work.
Prakerana Learn the data science process Grok the approach to
problem solving with data - Frame, Acquire, Refine, Transform, Explore, Model & Insight.
Abhyaas Practice, Practice, Practice. Repeat the data science process multiple
times on different types of problems.
Ruchi Do projects that interest you. Pick from work domain,
sports, economics, public policy etc. You are more likely to stick through the arduous journey of doing it.
Pradarshan Create a public profile. Share what you do so
people can connect with you. Create a public portfolio on Github or a blog or website.
Pratikriya Get Feedback. Put your work out there. It could
be a small analysis, visualisation, a mini-library. Nothing is too small to share.
Rachit Create a Data Science Portfolio. Work on at least
6 - 8 different projects over the next 1 year to build a portfolio of projects across topics.
Laaya Find your own rhythm. Build and do something every
day, week or month. Find your own cadence and start creating.
Advice for Beginners
None
Amit Kapoor @amitkaps amitkaps.com Bargava Subramanian @bargava