atinesh-s / coursera-machine-learning-stanford Goto Github PK

Are some of the assignments in the course peer graded? And getting stuck if someone picks up a snippet or two would he be marked for plag?

In nnCostFunction, when trying to calculate J = (1/m) * sum ( sum ( (-Y) .* log(output) - (1-Y) .* log(1-output) )); I get an error that says: Matrix dimensions must agree.

gradientDescent
Not enough input arguments.

Error in gradientDescent (line 5)
m = length(y);

Hello,
I am using your GitHub repository for submitting the assignment from the above-mentioned course. Now it is convenient to work in python for machine learning, I am trying to directly submit the answers from your repository using octave.
But I am facing the below issue while submitting the assignment.

"!! Submission failed: unexpected error: urlread: HTTP response code said er"

I would be grateful if I get help with the above issue.
Thanks and regards,
Dhanraj Mane

First of all, thanks for your effort. It would be nice if you provide the steps to getting the answers.

For example, Quiz 4 -- First question

The value of theta computed is [ 340412.65957447 109447.79558639 -6578.3539709 ] which means the more bedroom you have, the lower the price. I find this kinda weird, is it true? Also I tried to compute some samples from the dataset and the difference is kinda big (about 10%).

May there be any problems with the code?

Quiz 11.pdf in week 6 contains 1,2,3 and 5 but no 4th question.

Hey.... Thanks for everything you've been doing.... It really helps....... But I have a little issue.... It keeps telling me that there is a syntax error in the for loop

Hi, Still getting error,

gradientDescent
Not enough input arguments.

Error in gradientDescent (line 7)
m = length(y); % number of training examples

X_norm = (X - (t * mu)) ./ (t * sigma); %vectorized

In above Octave code, if X is 97 x 3 matrix then will it implicitly get converted to column vector?

Hey !

Could you show how you got the answer for week 2 quiz 4 as 0.32 ?

Thanks in advance.

1. In quiz 8, question 4, why answer B is not correct.

2. In assignment 2, For calculating costFunctionReg, I have done

h = sigmoid(X*theta);
theta_temp = [0; theta(2:end)];
J = (1/m)* sum((-y'*log(h)) - ((1-y)'*log(1-h))) + (lambda / (2*m)) * sum(theta .^ 2);
grad = (1/m)* sum((h-y)*X) + (lambda/m).*theta_temp;

My J is calculated as expected but getting error in grad. My code is same as like derivation (as I am seeing) but still unable to fig out where I am wrong. Whereas you are not using sum func but still your code is working fine. Can you pls explain how it is working?

Thanks for all of these.

Question, what other online course do you reccomend?

Greetings from Argentina.