Here I'll collect questions and give succint answers to possible questions in a machine learning interview. Use these only to check your understanding.
https://blog.mattbowers.dev/xgboost-from-scratch
System design
- ml system design
- Following manning books dist ml,design deep sys,churn-predic
General guide for ML interviews
- Summary/Notes of ML/DL
- AppliedAi notes: Useful for knowing common interview questions for each algorithm.
- Great Flash cards covering ML and DL
- CS 229 STANFORD
- Emma Ding's Youtube Channel
-
Anatomy of an ML interview
This are the algorithms that you should aim to know inside out. (Focus on Depth)
-
Background (Probability,Mult-calculus,Lin-algebra)
-
Decision Trees and Random Forests
Used Resources, jeremy howard decision tree/rf, code for fastai ml
General strategy: Once you are comfortable with the basics, take up SQL questions from stratscratch that have video solutions, try them by yourslef and them compare. Do them both in Python and MYSQL.
To learn refer these two courses
Learsql a. Basic SQL b. Window Functions
- Practice most commonly asked questions from stratscratch
- Video walk-throughs
- Practice SQL questions
- This educative course might be helpful
Focus on improving understanding, use brilliant.org
Recommended Courses:
- intro to prob(10),Perplexing probability(19),knowledge and uncertainty(13)
- Into to stat(26), stat 1(35), rand var n dist(28) Above courses together have about 130 lessons. Aiming about 5 lessons a day => 1 month. Good time. Suggested order: Intro to prob, Intro to stat, stat 1.
Extra: lin algebra with app (41),multivariable calc(44)
Some notes for devops.
Sampling theory(http://home.iitk.ac.in/~shalab/course1.htm) Design of exps(https://home.iitk.ac.in/~shalab/spanova.htm)
general