In the formulation, under some assumptions, we have obtained the perturbation formula, using which we can obtain z’ from z, as can be seen in the below image.
However, if we consider the expression, in the denominator, we have [1 - (w_g_T*w_t)^2]. Now, in the code, we have considered (w_g_T*w_t) as cos_theta, and therefore, (1 - cos_theta*cos_theta) forms the denominator. However, in the code, we have calculated sin_theta as np.sqrt(1 - cos_theta*cos_theta), so shouldn’t in the final expression of z_prime[j], we should use sin_theta^2? On the other hand, what we have used is simply sin_theta.
Have I done it wrong? Cause I don’t remember changing anything in this notebook! Thanks in advance!
Hey @mentor,
Can you take a look at this query?
Hi @Elemento, Thank you for the question. Our mentors will answer this shortly.
@aray, @Khushwanth, @SomeshChatterjee I would like to request you to take some time and answer this question.
Thank you,
Sharob
Hi Vishesh,
I went through your query and what you are pointing out seems logical to me so was waiting to see if anyone else knows about it. Let me double check else will take it up further.
Sorry for the delay in response.
Thanks a lot, @SomeshChatterjee. I will be looking forward to your reply 
Hi Vishesh,
I couldn’t figure out a reason why it should be that, I looked up the code with the original paper, the calculations are same there as well.
Let me see if any one else can clarify this. Sorry about this.
1 Like
No worries @SomeshChatterjee. Thanks a lot for your time. I will be looking forward to the justification 
