Continuous vs Discrete Variables in the context of Machine Learning.
Let’s get into the topic fast. I know, you don’t have time. You have to learn other topics too. Okay! I hear you :)
CONTINUOUS VARIABLE
A continuous variable can take any values. Think of it like this: If that number in the variable can keep counting, then its a continuous variable.
Ex: Weight of a person: 152.232 Kg, you’re probably thinking, “where am I counting?”. Yes, you are! The weight of the person is actually 152.232211223342211223332112244778899399947777889999888888377747666678788992336677……………………………………………………………………………………………………………………………………………………………………………………………………………………………………Kg
Obviously, those dots aren’t ending anytime soon. Matter of fact, they’re not ending!
Now you see how specifically that variable can “keep counting”? When I say “counting”, I’m referring to those counts after the decimal.
That is an example for continuous variable.
Can you think of another example?
Did you say Age? “You’re awesome!”. That’s correct! because age keeps counting. Don’t believe me? Install this and see for yourself.
Okay, other examples are time to train a deep neural network, income, cost of electricity, processing power of your brain (what!!??), your energy during night in J, just a quick remainder : J is the S.I unit of Energy.
I mean you know other examples now.
ML context : Continuous Variables are used for Regression.
DISCRETE/CATEGORIZED VARIABLE
A Discrete variable can take only a specific value amongst the set of all possible values or in other words, if you don’t keep counting that value, then it is a discrete variable aka categorized variable.
Example: Number of students in a university.
Think about Number of students in a university. Say a university has 75,123 students enrolled. Is that variable continuous?
You might say “yes” because, you’re intelligent. You’d suggest 75,123 is
75,123.000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000………………………………………………………………………………right?
Man! I too had the same thought, but let me tell you this: 75,123.00….=75,123, because, 0 after a decimal point doesn’t come into consideration at all. Haha. Statisticians are very intelligent! :) The point is, if the number is an integer (and obviously an integer doesn’t have decimals) then it is discrete.
Other examples:
The number of deep learning libraries in the market.
The number of GPU’s your regular computer has.
How good is your ML model: Say it only had 2 options {good, bad}. Then it’s a discrete variable.
Level of Agreement {Full, Partial, Not at all}
ML context : Discrete Variables are used for Classification.
To end this article let me ask you a question: If a variable A can take only values {22.3,225.69,122.23}, what kind of variable is it? why?
Continuous or Discrete? Don’t tell me “both”!
If you liked this article, then clap it up! :) Maybe a follow?
My social plug:
