Support Vector Machines (SVM) - the basics | simply explained
Вставка
- Опубліковано 14 сер 2022
- This video is intended for beginners
1. The equation of a straight line
2. The general form of a straight line (02:19)
3. The distance between a point and a line or between two lines (08:35)
4. The example data set (10:05)
5. How a SVM predicts new data (11:06)
6. The margin (13:25)
7. The classification (18:30)
8. How to maximize the margin (19:14)
9. The tuning parameter C (24:54)
10. How to evaluate the SVM (26:13)
This is by far the best explanation of SVM on youtube 👍
Agreed!!!
@@chiraglamba5435 totally agreed after checked more than 10 videos so far
Indeed... I'm really impressed with the explanation
The world need this type of lecturers
أنت أفضل من شرح SVM انت من عالم آخر، أقدر لك جهدك الواضح على المحتوى حتى أفضل من Ng ❤❤❤
one of the best explanations I've found for SVM. Thank you.
Hi very good job, I looked in several videos, the explanation Wt+x =1 (-1), you are the only one to explain it.Bravo
I 1000000000000000000 agree. Most places I looked they just gave the formula as if it is well known without any explanations.
no words...no one will get disappoint after seeing this video. great......... It would be difficult for me to understand svm if i didnt see your video
This is so wonderful. This video makes me so happy. The best explanation so far on SVM.
Finally understood the concept along with formula, thanks! 😀
Thank you so much, this is really helpful I finally understand now 24hr before my final
Wonderful. I needed to be led through that step by step. Thank you, thank you for making this video!
This really filled a big gap in my understanding in SVM. Thank you so much.
You saved my day. Thank you Andreas!
Wow, this is awesome and very insightful
I must subscribe to your channel
Thanks for your explanation. I really like the mathematical formulation you are presenting. Makes me wonder why math instructor didn't relate the abstract ideas to applicable ones like you are doing here. Keep it going please 👍
Thank you sooooo much. the best ever in explaining SVM!!
Absolutely Amazing on explanation of how SVM work! and Subscribed*!
Very insightful. Keep up the good work.
great explanation of preliminary materials for svm
Great video on SVM from just line equation
Thank you very much. Very well explained.
Thanks alot for this amazing video! Great explanation.
too good for explaining the svm with its formulae
Nice Explanation. Easy to grab 👏
best video according to my knowledge
Amazing explanation👍👍👍👍
Just awesome!!!!
beautifully explained
epic explanation!
God bless you man ❤
Excelent! Thank you
Thank you so much
Bro please make a complete video on ANN algorithm (regression and classification)
Artificial neural networks (ANN)
ua-cam.com/play/PLLTSM0eKjC2cSNGs87NyD3Q22t_qBEROY.html
How are you going to apply the SVM on the end
can you make a video about the dual form of SVM?
One question, how did you determine the red line at the beginning there(separating hyperplane)? Along with its equation, was that arbitrary? Did you eyeball it? Or is there some method to determine it?
I used a software, R, to get the optimal line. But if you would do this ”by hand”, you would just guess a line and then iterate to find the minimum of the cost function.
@@tilestats at 13.57 you said divide by 8 right. But in the equation after an equal sign we have zeros right. Then how are we getting -1 and +1?
good video
if we have yellow(3,3) intstead of yellow (3,4). How do we draw the straight line? how do we know the slope.
Can you make a separate video on kernel trick.
Greate
Hi this really is a very good video . Could you please explain why you multiply the equation by 4? Is it something that you always have to do in SVM? Thanks
No, 4 was just an arbitrary number to simplify the calculations by hand.
8:02👌
Bro how do we calculate the lines -4x+4y+-4=0 and -4x+4y+12=0 with y=x-1
You know that the lines should have a slope of 1, and that they should intercept the support vectors. Then you can calculate the intercepts of these lines, see for example here
www.wikihow.com/Find-the-Y-Intercept
We need the b (intercept) for the given points (3,4) to find this equation. Sub the values of x=3 and y=4 in the equation [-4x+4y+4=0], but instead of the intercept 4, take it as b. i.e.
sub the values of (3,4) in [-4x+4y+b=0]
-4.3 + 4.4 + b=0
-12 + 16 +b =0
b= -4
This -4 goes in the b of the [-4x+4y+b=0] and we get -4x+4y+ (-4)=0
Same steps for the next point (5,2):
Sub (5,2) in the equation [-4x+4y+b=0]
-4.5 + 4.2 + b =0
-20 + 8 + b =0
b = 12
Sub this b in the [-4x+4y+b=0] and we get -4x+4y+ (12)=0
Hope this helps
@@beenarizvi9777 how do we get -4 coefficient and +4 coefficient in -4x+4y + b=0
@@beenarizvi9777 well said :)
Anybody why does he swap the values of x and y?
Make a video on Naive bayes classifier
ua-cam.com/video/yRzIyWVEaCQ/v-deo.html