nb | Git versioning & syncing , Pandoc conversion

Below you will find my python code for a class assignment I was given a couple weeks ago which I have been unable to successfully debug. The problem is about finding the value at risk (i.e., the p% quantile) for an aggregate loss random variable, using FFT. We are given a clear mathematical procedure by which we can gain an estimation of the discretized CDF of the aggregate loss random variable. My results are, however, seriously off and I am making some kind of mistake which I have been unable to find even after hours of debugging my code.

The aggregate loss random variable S is given such that S=sum(X_i for i in range(N)), where N is negative binomially distributed with r=5, beta=.2, and X_i is exponentially distributed with theta=1. The probability generating function for this parametrization is P(z)=[1-\beta(z-1)]^{-r}.

We were asked to approximate the distribution of S by

1. choosing a grid width h and an integer n such that r=2^n is the number of elements to discretize X on,
2. discretizing X and calculating the probabilities of being in equally spaced intervals of width h,
3. applying the FFT to the discretized X,
4. applying the PGF of N to the elements of the Fourier-transformed X,
5. applying the inverse FFT to this vector.

The resulting vector should be an approximation for the probability masses of each such interval for S. I know from previous methods that the 95% VaR ought to be ~4 and the 99.9% VaR ought to be ~10. But my code returns nonsensical results. Generally speaking, my index where the ECDF reaches levels >0.95 is way too late, and even after hours of debugging I have not managed to find where I am going wrong.

I have also asked this question on the math stackexchange, since this question is very much on the intersection of programming and math and I have no idea at this moment whether the issue is on the implementation side of things or whether I am applying the mathematical ideas wrong.

I am working on a spatial search case for spheres in which I want to find connected spheres. For this aim, I searched around each sphere for spheres that centers are in a (maximum sphere diameter) distance from the searching sphere’s center. At first, I tried to use scipy related methods to do so, but scipy method takes longer times comparing to equivalent numpy method. For scipy, I have determined the number of K-nearest spheres firstly and then find them by cKDTree.query, which lead to more time consumption. However, it is slower than numpy method even by omitting the first step with a constant value (it is not good to omit the first step in this case). It is contrary to my expectations about scipy spatial searching speed. So, I tried to use some list-loops instead some numpy lines for speeding up using numba prange. Numba run the code a little faster, but I believe that this code can be optimized for better performances, perhaps by vectorization, using other alternative numpy modules or using numba in another way. I have used iteration on all spheres due to prevent probable memory leaks and …, where number of spheres are high.

I have 3 columns namely Models(should be taken as index), Accuracy without normalization, Accuracy with normalization (zscore, minmax, maxabs, robust) and these are required to be created as:

The present code selects minimum values by scanning the adjoining elements in the same and the succeeding row. However, I want the code to select all the values if they are less than the threshold value. For example, in row 2, I want the code to pick both 0.86 and 0.88 since both are less than 0.9, and not merely minimum amongst 0.86,0.88. Basically, the code should pick up the minimum value if all the adjoining elements are greater than the threshold. If that's not the case, it should pick all the values less than the threshold.

I'm looking for ways to count the number of trailing newlines from possibly binary data either:

• read from standard input
• or already in a shell variable (then of course the "binary" excludes at least 0x0) using POSIX or coreutils utilities or maybe Perl.

This should work without temporary files or FIFOs.

When the input is in a shell variable, I already have the following (possibly ugly but) working solution:

[Editing this question completely] Thank you , for those who helped in building the Periodic Table successfully . As I completed it , I tried to link it with another of my project E-Search , which acts like Google and fetches answers , except that it will fetch me the data of the Periodic Table .

But , I got a problem - not with the searching but with the layout . I'm trying to layout the x-scrollbar in my canvas which will display results regarding the search . However , it is not properly done . Can anyone please help ?

Below here is my code :

Just like the title says: The function to write has 2 parameters, the first is a number, the second a list of numbers:

example: (7, [5,3,4,7]) The Python function to write should return a list of numbers that when added would lead to '7' for example [3,4] or [4,3] or [7]

I wrote a Python function that actually works (see below bestSum()), but it only returns 1 working combination and I wish that someone could help edit it so it will return ALL possible "good" combinations. Then I could select the shortest one.

The recursive Python function is using a binary tree to go down by substracting a number from the list to the main number (substracting the same number is ok). If last node has 0 in it then that is a winner but if comes out as negative then that it is an impossible combination and should be dropped.

So instead of just returning, let say [4,3], it would return instead: ([3,4], [4,3], [7]) OR at least help me change the looping so it will not stop after it find the first working combination.

I have a joint probability density f(x,y,z) and I wish to find the conditional distribution X|Y=y,Z=z, which is equivalent to treating x as data and y and z as parameters (constants).

For example, if I have X|Y=y,Z=z being the pdf of a N(1-2y,3z^2+2), the function would be:

The following code works on "everyones" machine, except for mine: