Excellent video. Wonderfully explained. What I gathered from it is as follows: 1. Whenever you make an assumption of the function of your data, then it is a parametric ML algorithm (Linear Regression). However, if you do not make any assumptions about your function, then it is a non-parametric ML algorithm. So if you try to find out y=f(x), then it is parametric ML algorithm. 2. If the number of parameters does not grow with respect to the number of rows present in your data, then it is a parametric algorithm. However, if the number of parameters grow with respect to the number of rows present in your data, then it is a non-parametric algorithm. 3. It is incorrect to assume that non-parametric algorithms do not have parameters. It is just that they change or rather grow with the number of rows in your data.
Parameter are the The Nodes or split point of each feature. take example of cgpa and job_selection: If node is cgpa and it says if cgpa is greator than 5 , you will get job else you will not so on point 5 , consider a line below which all failed and obove which all passed. This is line, again when we go to higher dimension, line becomes plane and more
Excellent video. Wonderfully explained. What I gathered from it is as follows:
1. Whenever you make an assumption of the function of your data, then it is a parametric ML algorithm (Linear Regression). However, if you do not make any assumptions about your function, then it is a non-parametric ML algorithm. So if you try to find out y=f(x), then it is parametric ML algorithm.
2. If the number of parameters does not grow with respect to the number of rows present in your data, then it is a parametric algorithm. However, if the number of parameters grow with respect to the number of rows present in your data, then it is a non-parametric algorithm.
3. It is incorrect to assume that non-parametric algorithms do not have parameters. It is just that they change or rather grow with the number of rows in your data.
Wonderful explanation... keep doing what you are doing.
Your explanation is amazing sir
Please continue this playlist
Thanks for such a simplistic explanation ❤
Please continue tutorials in English. You explained it flawlessly!!
Amazing explanation. Thank you so much.... Linear regression is parametric then what about multiple linear regression?
Thank you sir provided by some knowledge
Great great explanation
great explaination sir
best explanation 👍
Bestest explanation
sir, when you are going to launch other question answers
Best!
thanks for the great explanation!!
Super Explaination
Thank you sir for this playlist
Nice
what are parameters for decision trees? just like slope and interceptor for linear regression.
Parameter are the The Nodes or split point of each feature. take example of cgpa and job_selection:
If node is cgpa and it says if cgpa is greator than 5 , you will get job else you will not
so on point 5 , consider a line below which all failed and obove which all passed.
This is line, again when we go to higher dimension, line becomes plane and more
thankyou sir
Hindi