This is implementation of Linear regression from scratch in NumPy only. In-depth explanation of key concepts like Cost Function and Gradient Descent kzbin.info/www/bejne/rammgquQgNRnnrc
@ShahramDerakhshandeh-sf7ldАй бұрын
That's a great.❤
@DeltaJes-co8yuАй бұрын
I cannot follow the accent unfortunately and even the CC is not working
@patriots7400Ай бұрын
why you shorten your last name? I want cite you!
@aihsdiaushfiuhidnva4 ай бұрын
This is very good! But where did you get Andrew's presentation?
@ontarioinctransport89125 ай бұрын
First comment enjoy
@adityavardhanjain5 ай бұрын
I wonder how the complexity of the model might affect the overfitting (or underfitting?)
@khoaphamquocanh49066 ай бұрын
Where can I watch this old course? Thanks
@betafishie10 ай бұрын
first
@sharedhardware10 ай бұрын
ㅤ
@ryanwang969910 ай бұрын
Great video!
@abdelrahmane657 Жыл бұрын
Thank you so much. It’s been very useful. 🙏👏
@helenareveillere3383 жыл бұрын
Hello, Do you know if I could listen to the sound of the MANIAC somewhere on the internet ? I'm a sound editor working on an audio documentary about Mathematics and litterature, and I need to recreate the sound of the MANIAC. Thanks for your answer. Helena
@AnTran-ot3qk3 жыл бұрын
The great video, thank you so much professor
@shivani404sheth43 жыл бұрын
so nicely explained. thank you!
@reachDeepNeuron3 жыл бұрын
instead of using superscript and subscript terms , had it been explained like start with the gist of what this algorithm does and then using math plus superscript , would help holding the audience and also motivating the audience to continue watching
@shahadp38683 жыл бұрын
Nicely done it sir...what about one vs one
@akashprabhakar63534 жыл бұрын
I did not get one thing...Suppose for a classification we get the max probability..then we wd be classifying only one class separately and rest 2 as another...but how are we classifying all 3 separately??
@samueldushimimana38314 жыл бұрын
well done Andrew
@nawabengineering43884 жыл бұрын
Well explained but Why is it called cost function? And taking 1/2 is not clear. Why and why not take square root?
@ditdit-dahdah-ditdit-dah3 ай бұрын
Cost function is also called as Loss function, both are synonyms. Division by m or 2m is interchangeable. What we are really concerned about it a model which is producing least error , not the value of the loss function directly. Cost functions can be of 3-types , among them is a regression, which again has 3 types , that is Mean Error , Mean squared error , mean absolute error . Why so many ? Cause a data set may have negative/positive errors , taking mean directly may cancel out +/- errors , and taking a square directly can be a bit troublesome if you have some outliers . In these videos , Andrew can be seen using all three in regression based . Note :it's not the only required param for concluding a model isn't good.
@nawabengineering43884 жыл бұрын
Everybody in this ML field directs to use python, you are the first one who referred to octave. Why is this so?
@jackmcgaughey43883 жыл бұрын
I know right
@elbrenantonio52564 жыл бұрын
Any video for multiclass entropy and entropy. Please show calculations sample. Thanks.
@bismeetsingh3524 жыл бұрын
Don't you have legal issues for copying content from coursera
@thesteve03454 жыл бұрын
I am pretty sure coursera copied from his content.
@GelsYT4 жыл бұрын
he is coursera
@jaideepsingh79553 жыл бұрын
@@GelsYT hahaha true..
@randomcowgoesmoo35464 жыл бұрын
Thanks Andrew Yang, I'll definitely vote for you.
@LouisDuran2 ай бұрын
wrong dude, the other guy wants to give you UBI. This guy wants to give you OVA
@ZombieLincoln6664 жыл бұрын
audio quality is shit
@swathys78184 жыл бұрын
Thank you for great explanation Sir!
@sanketneema2864 жыл бұрын
thankyou sir
@truettbloxsom84844 жыл бұрын
Just wanted to say these videos are amazing! thank you!
@dream1919194 жыл бұрын
There is an error of the example Andrew used here to demonstrate Normal equation. The X is the 4 by 5 matrix which makes the system underdetermined, which also result in the inverse of X's transpose multiplying X having no inverse. So the Normal equation cannot be calculable.
@bonipomei3 жыл бұрын
X is 4x5 and X(transpose) is 5x4. Therefore, X(transpose)*X = 5x4 * 4x5 which results in a 5x5 matrix, which has an inverse.
@IamPdub5 жыл бұрын
Great video, can you make a video on Stemming with Multiclass Classification?
@heller41965 жыл бұрын
Get this man a good Camera and Mic.
@namangupta86093 жыл бұрын
Sending you the bill...
@abdelrahmane657 Жыл бұрын
@@namangupta8609 Did you receive the bill? Or you’ll be the only youtuber watching this video
@ashwiniabhishek15045 жыл бұрын
Great video
@punkntded5 жыл бұрын
What does theta represent?
@ofathy19815 жыл бұрын
learning rate
@ByteSizedBusiness5 жыл бұрын
@@ofathy1981 alpha is the learning rate in gradient descent .... theta is a parameter like weights in NN
@MelvinKoopmans5 жыл бұрын
@@ofathy1981 Theta does not represent the learning rate, instead it represents the parameters of the model (e.g. the weights). So P(y | x; θ) translates to English as "The probability of *y* given *x* , parameterized by *θ* ".
@amirdaneshmand97432 жыл бұрын
That the parameters of logistic classifier which is trained separately for each case