I have 40 variables. The 40 variables names are in a cell array (40 x 1).
Each variable will have a matrix. The matrix is of type of double and will be size 5000 x 150. It will also have a vector of size 1 x 150 & one last vector 1 x 4.
Initially I was going to dynamically name each struct with its variable name in the cell array. So would look like something like below (assuming variable name is ABC),
ABC.dataMatrix
ABC.dataVec
ABC.summaryData
All the variables would be saved to a directory.
However I've read using eval isn't a very good idea so guessing my idea isn't the best way to go about this problem. What would be the best way to go about this problem?
You can either use struct arrays with dynamic field names, as #Shai and #RobertStettler suggest.
However, another option is a table. This might be more appealing if you want to see your data in one big matrix, and you can give each table row the name of your variables too! Note that the rows in a table would then be what you call your variables, but MATLAB calls the table columns its variables.
Also note that using a table can be more difficult than using struct or cell arrays, but if you know how to use them, you can handle a table too.
An example:
% create dummy data
rowNames = {'a';'b';'c'};
M = {ones(3); zeros(3); nan(3)}; % a cell, one element per item in rowNames
V = [1 2; 3 4; 5 6]; % a matrix of vectors, each row containing a vector for every item in rowNames
% create a table
T = table(M,V,'RowNames',rowNames); % this is where your variable names could go
Now, to access data you could use (some examples):
T(2,:) or T('b',:), return a table for all data on the 'b' row.
T(:,2) or T(:,'V'), return a table of variable V for all rows.
T.V or T{:,2} or T{:,'V'} or T.(2), return matrix V for all rows. This syntax is similar to accessing a (dynamic) field name of a struct.
T{3,1} or T{'c',1} or T.M('c'), return cell M for row 'c'. This syntax is similar to accessing a cell, but with more advanced possibilities, i.e. the ability to access the table via row or variable names.
T{3,1}{:} or T{'c',1}{:} or T.M{'c'}, return cell contents M for row 'c'.
And even more complex: T('a',:).M{:} is a complex way of accessing the cell content of M for row 'a', which can be done with T{1,1}{:} or T.M{'a'} or T{'a','M'}{:} or T.M{1} as well.
In your case you would en up with a 40x3 table, with every row what you call a variable and the first column the matrices (in cell arrays), and the last two columns the vectors (as well in cell arrays or as a 40xm double, m being the length of your vector).
Related
I have a row cell vector M, containing matrices in each cell. Every matrix m (matrix inside the big matrix M) is made of 2 channels (columns), of which I only want to use the first.
The approach I thought about was going through each m, check if it has 2 channels, and if that is the case delete the second channel.
Is there a way to just slice it in matlab? or loop it and obtain the matrix M as the matrix m would disappear.
First code is:
load('ECGdata.mat')
I have the below.
when I double-click in one of the variable , here is what I can see:
As you can see the length of each matrix in each cell is different. Now let's see one cell:
The loop I'm trying to get must check the shape of the matrix (I'm talking python here/ I mean if the matrix has 2 columns then delete the second) because some of the variables of the dataframe have matrix containing one column (or just a normal column).
In here I'm only showing the SR variable that has 2 columns for each matrix. Its not the case for the rest of the variables
You do not need to delete the extra "channel", what you can do is quite simple:
newVar = cellfun(#(x)x(:,1), varName, 'UniformOutput', false);
where varName is SR, VF etc. (just run this command once for each of the variables you load).
What the code above does is go over each element of the input cell (an Nx2 matrix in your example), and select the first column only. Then it stores all outputs in a new cell array. In case of matrices with a single column, there is no effect - we just get the input back.
(I apologize in advance if there is some typo / error in the code, as I am writing this answer from my phone and cannot test it. Please leave a comment if something is wrong, and I'll do my best to fix it tomorrow.)
I'm really sorry to bother so I hope it is not a silly or repetitive question.
I have been scraping a website, saving the results as a collection in MongoDB, exporting it as a JSON file and importing it in MATLAB.
At the end of the story I obtained a struct object organised
like this one in the picture.
What I'm interested in are the two last cell arrays (which can be easily converted to string arrays with string()). The first cell array is a collection of keys (think unique products) and the second cell array is a collection of values (think prices), like a dictionary. Each field is an instance of possible values for a set of this keys (think daily prices). My goal is to build a matrix made like this:
KEYS VALUES_OF_FIELD_1 VALUES_OF_FIELD2 ... VALUES_OF_FIELDn
A x x x
B x z NaN
C z x y
D NaN y x
E y x z
The main problem is that, as shown in the image and as I tried to explain in the example matrix, I don't always have a value for all the keys in every field (as you can see sometimes they are 321, other times 319 or 320 or 317) and so the key is missing from the first array. In that case I should fill the missing value with a NaN. The keys can be ordered alphabetically and are all unique.
What would you think would be the best and most scalable way to approach this problem in MATLAB?
Thank you very much for your time, I hope I explained myself clearly.
EDIT:
Both arrays are made of strings in my case, so types are not a problem (I've modified the example). The main problem is that, since the keys vary in each field, firstly I have to find all the (unique) keys in the structure, to build the rows, and then for each column (field) I have to fill the values putting NaN where the key is missing.
One thing to remember you can't simply use both strings and number in one matrix. So, if you combine them together they can be either all strings or all numbers. I think all strings will work for you.
Before make a matrix make sure that all the cells have same element.
new_matrix = horzcat(keys,values1,...valuesn);
This will provide a matrix for each row (according to your image). Now you can use a for loop to get matrices for all the rows.
For now, I've solved it by considering the longest array of keys in the structure as the complete set of keys, let's call it keys_set.
Then I've created for each field in the structure a Map object in this way:
for i=1:length(structure)
structure(i).myMap = containers.Map(structure(i).key_field, structure(i).value_field);
end
Then I've built my matrix (M) by checking every map against the keys_set array:
for i=1:length(keys_set)
for j=1:length(structure)
if isKey(structure(j).myMap,char(keys_set(i)))
M(i,j) = string(structure(j).myMap(char(keys_set(i))));
else
M(i,j) = string('MISSING');
end
end
end
This works, but it would be ideal to also be able to check that keys_set is really complete.
EDIT: I've solved my problem by using this function and building the correct set of all the possible keys:
%% Finding the maximum number of keys in all the fields
maxnk = length(structure(1).key_field);
for i=2:length(structure)
if length(structure(i).key_field) > maxnk
maxnk = length(structure(i).key_field);
end
end
%% Initialiting the matrix containing all the possibile set of keys
keys_set=string(zeros(maxnk,length(structure)));
%% Filling the matrix by putting "0" if the dimension is smaller
for i=1:length(structure)
d = length(string(structure(i).key_field));
if d == maxnk
keys_set(:,i) = string(structure(i).key_field);
else
clear tmp
tmp = [string(structure(i).key_field); string(zeros(maxnk-d,1))];
keys_set(:,i) = tmp;
end
end
%% Merging without duplication and removing the "0" element
keys_set = union_several(keys_set);
keys_set = keys_set(keys_set ~= string(0));
Lets say i have a matrix A of 300x65. the last column(65th) contains ordered values (1,2,3). the first 102 elements are '1', the second 50 elements are '2' and the remainder will be '3'.
I have another matrix B, which is 3x65 and i want to copy the first row of B by the number of '1's in matrix A. The second row of B should be copied by the number of '2's in in matrix A and the 3th row should be copied by the remaining value of matrix A. By doing this, matrix B should result in a 300x65 matrix.
I've tried to use the repmat function of matlab with no succes, does anyone know how to do this?
There are many inconsistencies in your problem
first if you copy 1 row of B for every element of A(which will end up happening by your description) that will result in a matrix 19500x65
secondly copy its self is a vague term, do you mean duplicate? do you want to store the copied value into a new var?
what I gathered from your problem is you want to preform some operation between A and B to create a matrix and store it in B which in itself will cause the process to warp as it goes if you do not have another variable to store the result in
so i suggest using a third variable c to store the result in and then if you need it to be in b set b = C
also for whatever process you badly described I recommend learning to use a 'for' loop effectively because it seems like that is what you would need to use
syntax for 'for' loop
for i = [start:increment:end]
//loops for the length of [start:increment:end]
//sets i to the nth element of [start:increment:end] where n is the number of times the loop has run
end
If I understand your question, this should do it
index = A(:,end); % will be a column of numbers with values of 1, 2, or 3
newB = B(index,:); % B has 3 rows, which are copied as required by "index"
This should result in newB having the same number of rows as A and the same number of columns as the original B
i've got a cell array full of numbers, with 44 rows and different column length in each row
how could i calculate the number of columns in each row?(the columns which their contents are not empty)
i've used 2 different ways which both of them where wrong
the 1st one:
%a is the cell array
s=length(a)
it gives 44 which is the number of rows
the 2nd one
[row, columms]=size(a)
but it doesn't work either cause the number of columns is different in each row.
at least i mean the number of columns which are not empty
for example i need the number of columns in row one which it is 43(a{1 1:43}) but it gives the number of columns for each elements like a{1,1} which is 384 or a{1,2},a{1,3} and so on
You need to access each member of the cell array separately, you are looking for the size of the data contained in the cell - the cell is the container. Two methods
for loop:
cell_content_lengths=zeros(1,length(a));
for v=1:length(a)
cell_content_lengths(v)=length(a{v});
end
cellfun:
cell_content_lengths=cellfun(#length,a);
Any empty cells will just have length 0. To extend the for-loop to matrices is trivial, and you can extend the cellfun part to cells containing matrix by using something like this, if you are interested:
cell_content_sizes=cell2mat(cellfun(#length,a,'uniformoutput',false));
(Note for the above, each element of a needs to have the same dimension, otherwise it will give errors about concatenating different size matrices)
EDIT
Based on your comment I think I understand what you are looking for:
non_empty_cols = sum(~cellfun(#isempty,a),2);
With thanks to #MZimmerman6 who understood it before me.
So what you're really asking, is "How many non-empty elements are in each row of my cell array?"
filledCells = ~cellfun(#isempty,a);
columns = sum(filledCells,2);
I'm currently working in an area that is related to simulation and trying to design a data structure that can include random variables within matrices. To motivate this let me say I have the following matrix:
[a b; c d]
I want to find a data structure that will allow for a, b, c, d to either be real numbers or random variables. As an example, let's say that a = 1, b = -1, c = 2 but let d be a normally distributed random variable with mean 0 and standard deviation 1.
The data structure that I have in mind will give no value to d. However, I also want to be able to design a function that can take in the structure, simulate a uniform(0,1), obtain a value for d using an inverse CDF and then spit out an actual matrix.
I have several ideas to do this (all related to the MATLAB icdf function) but would like to know how more experienced programmers would do this. In this application, it's important that the structure is as "lean" as possible since I will be working with very very large matrices and memory will be an issue.
EDIT #1:
Thank you all for the feedback. I have decided to use a cell structure and store random variables as function handles. To save some processing time for large scale applications, I have decided to reference the location of the random variables to save time during the "evaluation" part.
One solution is to create your matrix initially as a cell array containing both numeric values and function handles to functions designed to generate a value for that entry. For your example, you could do the following:
generatorMatrix = {1 -1; 2 #randn};
Then you could create a function that takes a matrix of the above form, evaluates the cells containing function handles, then combines the results with the numeric cell entries to create a numeric matrix to use for further calculations:
function numMatrix = create_matrix(generatorMatrix)
index = cellfun(#(c) isa(c,'function_handle'),... %# Find function handles
generatorMatrix);
generatorMatrix(index) = cellfun(#feval,... %# Evaluate functions
generatorMatrix(index),...
'UniformOutput',false);
numMatrix = cell2mat(generatorMatrix); %# Change from cell to numeric matrix
end
Some additional things you can do would be to use anonymous functions to do more complicated things with built-in functions or create cell entries of varying size. This is illustrated by the following sample matrix, which can be used to create a matrix with the first row containing a 5 followed by 9 ones and the other 9 rows containing a 1 followed by 9 numbers drawn from a uniform distribution between 5 and 10:
generatorMatrix = {5 ones(1,9); ones(9,1) #() 5*rand(9)+5};
And each time this matrix is passed to create_matrix it will create a new 10-by-10 matrix where the 9-by-9 submatrix will contain a different set of random values.
An alternative solution...
If your matrix can be easily broken into blocks of submatrices (as in the second example above) then using a cell array to store numeric values and function handles may be your best option.
However, if the random values are single elements scattered sparsely throughout the entire matrix, then a variation similar to what user57368 suggested may work better. You could store your matrix data in three parts: a numeric matrix with placeholders (such as NaN) where the randomly-generated values will go, an index vector containing linear indices of the positions of the randomly-generated values, and a cell array of the same length as the index vector containing function handles for the functions to be used to generate the random values. To make things easier, you can even store these three pieces of data in a structure.
As an example, the following defines a 3-by-3 matrix with 3 random values stored in indices 2, 4, and 9 and drawn respectively from a normal distribution, a uniform distribution from 5 to 10, and an exponential distribution:
matData = struct('numMatrix',[1 nan 3; nan 2 4; 0 5 nan],...
'randIndex',[2 4 9],...
'randFcns',{{#randn , #() 5*rand+5 , #() -log(rand)/2}});
And you can define a new create_matrix function to easily create a matrix from this data:
function numMatrix = create_matrix(matData)
numMatrix = matData.numMatrix;
numMatrix(matData.randIndex) = cellfun(#feval,matData.randFcns);
end
If you were using NumPy, then masked arrays would be the obvious place to start, but I don't know of any equivalent in MATLAB. Cell arrays might not be compact enough, and if you did use a cell array, then you would have to come up with an efficient way to find the non-real entries and replace them with a sample from the right distribution.
Try using a regular or sparse matrix to hold the real values, and leave it at zero wherever you want a random variable. Then alongside that store a sparse matrix of the same shape whose non-zero entries correspond to the random variables in your matrix. If you want, the value of the entry in the second matrix can be used to indicate which distribution (ie. 1 for uniform, 2 for normal, etc.).
Whenever you want to get a purely real matrix to work with, you iterate over the non-zero values in the second matrix to convert them to samples, and then add that matrix to your first.