|
@iamtrask | |||||
|
Machine Learning in a company is 10% Data Science & 90% other challenges
It's VERY hard. Everything in this guide is ON POINT, and it's stuff you won't learn in an ML book
"Best Practices of ML Engineering"
This is a lifesaver #100DaysOfMLCode project
martin.zinkevich.org/rules_of_ml/ru… pic.twitter.com/eYJ5KpQmVt
|
||||||
|
||||||
|
Parisa Rashidi
@Parisa__Rashidi
|
12. pro |
|
I would say this is also true in academia if working on applied problems. For example, ML applications in healthcare: a lot more beyond a robust analysis, including privacy issues, human subjects, knowledge of regulations, deployment ecosystem, etc.
|
||
|
|
||
|
MegaExponential TF
@furlanel
|
12. pro |
|
Rule 33 is simply wrong and introduce info-leakage from test to train for non iid data.
|
||
|
|
||
|
Dieter Castel
@DieterCastel
|
12. pro |
|
Very true. Also my (albeit limited) experience. It's not magic, it's still software and software engineering is hard. Secure software even harder.
|
||
|
|
||
|
Sidharth Ramesh
@sidharth_ramesh
|
12. pro |
|
Marking it for later
|
||
|
|
||
|
Imran Khawaja
@krazyimran
|
12. pro |
|
We are going through this right now and I agree 100. Figuring out what data you have and logging more is one of the first steps as well as how to store it.
|
||
|
|
||
|
Ayoub Benaissa
@y0uben11
|
12. pro |
|
Loved rule #1
|
||
|
|
||
|
Sandeep Sharma
@sandeepkaushik
|
12. pro |
|
Resonates so much ☝️
|
||
|
|
||
|
Aman sharma
@Aman73736425
|
12. pro |
|
So true
|
||
|
|
||
|
Vaibhav Kumar Gupta
@Vaibhav88406996
|
13. pro |
|
So true man ,😂
|
||
|
|
||