Linear Regression in 2 minutes
Вставка
- Опубліковано 8 вер 2024
- Linear Regression in 2 minutes.
---------------
Credit:
🐍 Manim and Python : github.com/3b1...
🐵 Blender3D: www.blender.org/
🗒️ Emacs: www.gnu.org/so...
Music/Sound: www.bensound.com
This video would not have been possible without the help of Gökçe Dayanıklı.
This is a very good exercise for people who already know the topic, but want to learn new ways of explaining it to others in a very short time.
Your videos are absolutely stunning, only matter of time when the serious views will come in. Please keep doing these! :)
Thanks for the great feedback. stay tuned for more!
Wow, you saved my life ! This is the most clear explanation about this topic. Loved it ! Love you !
Cap
Thank you SOOO much for this video.. My intergated 1 teacher was absent today and I didn't understand the topic. I then stumbled upon this video and it helped alot. THANKS!
Most awesome videos that explains things how they are suppose to be. Love it
00:01 Linear regression predicts the value of a dependent variable based on an independent variable.
00:20 Linear regression helps predict new values based on existing data
00:38 Linear regression is useful for prediction problems.
00:55 Quantifying the fit of a line to data points
01:10 Solving a prediction machine learning problem using linear regression from scratch.
01:29 Linear regression is simple and extensible
01:49 Linear regression allows interpretation of coefficients
02:08 Linear regression involves finding the relationship between two variables.
Great content! it's so helpful to see mathematical concept visually explained! Thank you for making it so easy and available.
This Channel is Really Creating some masterpieces in Visually Explaining Some Really deep Concepts in Optimization, A.I & Machine Learning....The Art with which you Visually Explain Things is Truly Phenomenal,
I've Been a Huge Fan of this Channel since DAY 1 ,Please Never Stop Creating this type of content ….. !!!
As an Optimization Student ,Is There Any Way I can connect with you and help this channel create more good Content..??
Hey there, thanks a lot for watching! I am glad you enjoyed my channel. Feel free to DM me on twitter.
Thank you. Perfectly simplified in 2 minutes. Now I can build on this basic understanding
Very well explained.. it helps to understand how the function does the job.. ignore negative comments..😅
Thank you so much for this amazing effort
simple and easy to understand.
Awesome video! I love how you explain every concept visually and easy to understand. Any chance on making a video on logistic regression as well?
Thank you sm for this simple explanation it helps a lotttt
This is the best way for video lectures
Amazing explanation and animations!
This was very helpful. Thank You
Amazing explnanation
What an actual boss of a video
Perfect video, thanks
You're welcome!
beautiful
Can someone please explain how the gradient is applied to the summation over the square error term to produce the result shown at the bottom at 1:13
this is masterpiece of information you deserve more views
Thanks for the amazing comment!
that was life saving TSM!!!
Just Amazing!
Best explanation of linear regression that I've heard.
Simple and nice thanks
Wooow great video!
What happened after 1:10, what was that gradient thing?
Thank you sir this one saves me
Please do Random Forest Regression and Knn Regression.
How did we get those equations to solve for alpha and beta?
Good video
Nice ! Amazing enough for a beginner like me!! @@
What is 'n' in the formula for beta at 1:18?
I assume it is the number of points but I can't get my code to work so I'd like to be sure.
Thanks
Update: after some testing it is clear that the issue is either with the formula for alpha or my implementation of it.
Everything works when beta is 0, but as i change my sample to have a value of beta further and further from 0 alpha gets offset more and more for some reason.
i am about to go through my third re-write of the thing and i don't know what to change.
if anyone wants to try and figure out what is wrong then here's some results:
Sample: alpha = 0.5, beta = 0 Result: alpha = ~0.5
Sample: alpha = 0.5, beta = 50 Result: alpha = ~0.25
Sample: alpha = 0.5, beta = 100 Result: alpha = ~0
yes the error does grow linearly with beta.
Update 2: re-wrote it for the third time. still wrong in the exact same way. send help.
The formula is wrong. the formula for alpha is wrong. dem = sum(x_i^2) - n*sum(x_i)^2, nom = n*sum(x_i * y_i) - sum(x_i)*sum(y_9)
thank you for the python code!
That's great!
1:20 the formula for alpha is wrong. dem = sum(x_i^2) - n*sum(x_i)^2, nom = n*sum(x_i * y_i) - sum(x_i)*sum(y_9)
loved it
Are you using manim for these presentations or something else ?
I am using a combination of manim, blender, and after effects.
@@VisuallyExplained Thanks
wow very nice can you explain pandas and dataframe with all methods I am Ai student
great video
only for those who already know it
I think I have a problem with the formula because when i try it to an example, it completely wrong.
I used linear regression in my forex bot!
wow 😊✨
Mehn, great content
am i stupid or maybe with the double equations your gave you can't find values of alpha and beta because one equation is just multple of the other
HOLY $H!T
I went from a 1 minute explanation video to a 2 minute one :/
ابوس راسك ياشيخ
It's not lazy it's time wasting, why would anyone waste their time doing something that a program can give answers to we just need to understand it.
This is not a good video who wants to learn all of this stuff from scratch!
It's 2 minutes long, I'm sorry but what did you expect?
I learned what I needed form this. Did you think you're gonna learn everything from scratch from a 2 minute vid?
I do! I needed this video!!!
I’d suggest- Statquest Josh Starmer. For anyone zero-hero
@@vishalvarakhedkar8816 I love his ukulele videos, lol
The formula is wrong.. he conveniently moved on...Read below many have commented on this....
please make your speech little slower so you can be heard without giving stress
Just learn how to speak slwoly and you will have more views
That "gradient" wasn't explained. And your a definition is missing two n*.
This is very poorly explained.