To plot one stem with multiple values for data of different months.

Dear MATLAB Experts,

I have data in .mat files, each corresponding to a particular month with selected_dates and brightness information.

Each month has a different number of output values (1, 2, 3, ..., N). I need to create a stem plot for each month, with the month on the x-axis and brightness on the y-axis.

For example, if D201701 has six values, the plot will show six values in one single stem, and the x-axis will display the month name.

Additionally, I want to annotate each stem value with a marker (e.g., . or *) to indicate the corresponding date. I have attached an image that shows the stem plot for multiple values, as well as the data and code.

Thank you for guidance and help.

[Cris LaPierre]

Are you trying to duplicate the image you shared? If so, I don't see how this is a stem plot, since stem plots would all originate from a common horizontal line. It looks like a bunch of line plots to me.

[Amjad Iqbal]

@Cris LaPierre

Yes, I actually wanted to emphasis that I want to plot one stem line for one month to show all the values corresponding to brightness. I understand from image its confusing.

@Voss understood the query and resolved. I appreicate both for time and efforts.

I will further norish the results.

Answer 1

[Voss]

Maybe something along these lines:

unzip('Processed_results.zip')

% load the files
F = dir(fullfile('Processed_results','*.mat'));
NF = numel(F);
for ii = 1:NF
    S = load(fullfile(F(ii).folder,F(ii).name));
    F(ii).brightness = S.all_brightness_values;
    F(ii).dates = S.selected_dates;
end

% count the number of dates and brightnesses in each file
N_dates = arrayfun(@(x)numel(x.dates),F);
N_brightness = arrayfun(@(x)numel(x.brightness),F);
assert(isequal(N_dates,N_brightness))

% collect date and brightness values into matrices;
% each column is a file, each row is a brightness/date
NB = max(N_brightness);
dates = NaT(NB,NF);
brightness = NaN(NB,NF);
for ii = 1:NF
    dates(1:N_dates(ii),ii) = F(ii).dates;
    brightness(1:N_brightness(ii),ii) = F(ii).brightness;
end

% construct datetime dt representing the start of each month
dt = datetime(year(dates(1,:)),month(dates(1,:)),1);

% markers to be used in plots
markers = '.*xo+^v<>sdph';
markers = repmat(markers,1,ceil(NB/numel(markers)));

% plot stems
figure('Position',[50 50 1500 600])
hold on
for ii = 1:NB
    stem(dt,brightness(ii,:),'Marker',markers(ii),'MarkerSize',12)
end
xticks(min(dt):calmonths(1):max(dt))

% or plot markers only
figure('Position',[50 50 1500 600])
hold on
for ii = 1:NB
    plot(dt,brightness(ii,:),'LineStyle','none','Marker',markers(ii),'MarkerSize',12)
end
xticks(min(dt):calmonths(1):max(dt))

[Amjad Iqbal]

@Voss I really appreciate your kind efforts and providing the solution.

This is what I wanted to plot and show the results. Once again Thank you.

[Voss]

You're welcome!

Answer 2

[Mathieu NOE]

hello @Amjad Iqbal

I modified a bit your code

added one line in the first for loop

    number_of_data_thismonth(k) = numel(brightnessValues);  % <= added this line first

then modified the plot

    % original plot
    % for i = 1:numel(allDates)
    %     % Plotting the stem for each date
    %     stem(allDates(i), allBrightnessValues(i), 'filled');
    %     % Extracting time information
    %     time_info = datestr(allDates(i), 'HH:MM:SS');
    %     % Adding time information on top of each stem
    %     text(allDates(i), allBrightnessValues(i), time_info, 'VerticalAlignment', 'bottom', 'HorizontalAlignment', 'center');
    % end
    
    % modified plot
    nnn = 1+[0 cumsum(number_of_data_thismonth)]; % cumulative sum of nb of data per month
    for i = 1:numel(number_of_data_thismonth)
        ind_start = nnn(i);
        ind_stop = nnn(i+1) - 1;
        ll(i) =  ind_stop - ind_start +1; % ll is equal to number_of_data_thismonth, so we have a working code 
        % Plotting the stem for each date
        % x position is "center" data 
        ind = round(mean(ind_start:ind_stop));
        stem(repmat(allDates(ind),ll(i),1), allBrightnessValues(ind_start:ind_stop), 'filled');
        % Extracting time information
        time_info = datestr(allDates(ind_start:ind_stop), 'HH:MM:SS');
        % Adding time information on top of each stem
        text(repmat(allDates(ind),ll(i),1), allBrightnessValues(ind_start:ind_stop), time_info, 'VerticalAlignment', 'bottom', 'HorizontalAlignment', 'center');
    end

your original plot looks like

now the modified plot is :

try it ; hope it helps - full code below

clear;clc;close;

% List all .mat files in the directory
matFiles = dir('*.mat');

% Initialize cell arrays to hold aggregated data
allBrightnessValues = [];
allDates = [];

% Loop through each .mat file
for k = 1:length(matFiles)
    % Load the current .mat file
    data = load(matFiles(k).name);
    
    % Extract the data
    brightnessValues = data.all_brightness_values;
    dates = data.selected_dates;
    
    % Ensure data consistency
    if isrow(brightnessValues)
        brightnessValues = brightnessValues';
    end
    
    number_of_data_thismonth(k) = numel(brightnessValues);  % <= added this line first
    
    % Convert dates to cell array of strings if necessary
    if ischar(dates)
        dates = cellstr(dates);
    elseif isdatetime(dates)
        dates = cellstr(datestr(dates, 'dd-mmm-yyyy HH:MM:ss'));
    elseif iscell(dates)
        dates = cellfun(@char, dates, 'UniformOutput', false);
    else
        error('Unexpected date format in file: %s', matFiles(k).name);
    end
    
    % Append the data to the aggregated vectors
    allBrightnessValues = [allBrightnessValues; brightnessValues];
    allDates = [allDates; dates];
end

% Convert the date strings to datetime format
allDates = datetime(allDates, 'InputFormat', 'dd-MMM-yyyy HH:mm:ss');

% Display the size of the aggregated vectors
disp(['Total number of brightness values: ', num2str(length(allBrightnessValues))]);
disp(['Total number of dates: ', num2str(length(allDates))]);


% Plotting
figure;
hold on;

% original plot
% for i = 1:numel(allDates)
%     % Plotting the stem for each date
%     stem(allDates(i), allBrightnessValues(i), 'filled');
%     % Extracting time information
%     time_info = datestr(allDates(i), 'HH:MM:SS');
%     % Adding time information on top of each stem
%     text(allDates(i), allBrightnessValues(i), time_info, 'VerticalAlignment', 'bottom', 'HorizontalAlignment', 'center');
% end

% modified plot
nnn = 1+[0 cumsum(number_of_data_thismonth)]; % cumulative sum of nb of data per month
for i = 1:numel(number_of_data_thismonth)
    ind_start = nnn(i);
    ind_stop = nnn(i+1) - 1;
    ll(i) =  ind_stop - ind_start +1; % ll is equal to number_of_data_thismonth, so we have a working code 
    % Plotting the stem for each date
    % x position is "center" data 
    ind = round(mean(ind_start:ind_stop));
    stem(repmat(allDates(ind),ll(i),1), allBrightnessValues(ind_start:ind_stop), 'filled');
    % Extracting time information
    time_info = datestr(allDates(ind_start:ind_stop), 'HH:MM:SS');
    % Adding time information on top of each stem
  text(repmat(allDates(ind),ll(i),1), allBrightnessValues(ind_start:ind_stop), time_info, 'VerticalAlignment', 'bottom', 'HorizontalAlignment', 'center');
end

hold off;
title('Brightness Temperature vs. Date');
xlabel('Date');
ylabel('Brightness Temperature');
datetick('x', 'yyyy-mm-dd', 'keeplimits');
ylim([180 280])




%% 
%Plotting
% figure;
% hold on;
% for i = 1:numel(allDates)
%     % Plotting the stem for each date
%     stem(allDates(i), allBrightnessValues(i), 'filled');
%     % Extracting time information
%     time_info = datestr(allDates(i), 'HH:MM:SS');
%     % Adding time information on top of each stem
%     text(allDates(i), allBrightnessValues(i), time_info, 'VerticalAlignment', 'bottom', 'HorizontalAlignment', 'center');
% end
% hold off;
% title('Brightness Temperature vs. Date');
% xlabel('Date'); axis normal;
% ylabel('Brightness Temperature');
% datetick('x', 'yyyy-mm-dd', 'keeplimits');

[Amjad Iqbal]

@Mathieu NOE

I really appreciate the time and efforts to provide the solution. I have modified the plot and I wanted to plot a neat one. Attached one is the finest result. I am thank full for all the experts to respond promptly.

I always get benifit from this platform to learn new ideas and adapt for several solutions.

[Mathieu NOE]

you're welcome !!

nice plot BTW :)