A repository comprising multiple functions for making elegant publication-quality figures in MATLAB
A repository comprising multiple functions for making elegant publication-quality figures in MATLAB.
figure_boxplot.m
)
generatePDF.m
)
figure_heatmap.m
)
Generating boxplot in MATLAB using the default function boxplot.m
is a bit cumbersome due to the large number of required (and somewhat strict in terms of format) inputs. Here, I have written a wrapper code for making nice boxplots quickly and efficiently.
Function: figure_boxplot.m
% Number of intended boxes in the figure
num_boxes = 8;
% Generating random data
data = cell(1,num_boxes);
for k = 1:num_boxes
data{k} = randi(10) + randn(1,1000);
end
% Using the "figure_boxplot.m" function to plot the boxplot figure using the data
figure_boxplot(data)
% Number of intended boxes in the figure
num_boxes = 6;
% Generating random data
data = cell(1,num_boxes);
for k = 1:num_boxes
data{k} = randi(10) + randn(1,1000);
end
% Using the "figure_boxplot.m" function to plot the boxplot figure using the data,
% x- and y-axis labels, and label of each box.
% For more information related to function inputs, check the function "figure_boxplot.m"
label_axes = {'Variable','Number'};
label_boxes = {'alpha','beta','gamma','delta','epsilon','zeta'};
figure_boxplot(data,label_axes,label_boxes);
Instead of a cell, the code works even if the input is in matrix form of size num_samples x num_boxes.
% Number of intended boxes in the figure
num_boxes = 7;
% Number of samples in each box plot
num_samples = 1000;
% Generating random data
data = zeros(num_samples,num_boxes);
for k = 1:num_boxes
data(:,k) = randi(10) + randn(num_samples,1);
end
% Using the "figure_boxplot.m" function to plot the boxplot figure using the data
figure_boxplot(data)
This is a code to generate nice (properly normalized) probability density function (PDF) plots with minimum amount of input arguments.
Function: generatePDF.m
% Generating random data
x = randn(1,500);
figure;
generatePDF(x)
% Loading colors
run colors_definitions.m
% Generating random data
x = randn(1,500);
figure;
subplot(2,1,1)
generatePDF(x,'b') %You can use any color
title('Using one of the default colors')
subplot(2,1,2)
generatePDF(x,color_scheme_set1(1,:)) %You can use any color
title('Specifying a color manually')
% Loading colors
run colors_definitions.m
% Generating random data
x = randn(1,5000);
figure;
subplot(3,1,1)
generatePDF(x,color_scheme_set1(1,:),'hist')
title('Histogram plot')
subplot(3,1,2)
generatePDF(x,color_scheme_set1(1,:),'curve')
title('Curve plot')
subplot(3,1,3)
generatePDF(x,color_scheme_set1(1,:),'area')
title('Area plot')
% Loading colors
run colors_definitions.m
% Generating random data
x = randn(1,5000);
% Specifying the number of bins
no_of_bins = [20 30 50];
figure;
for k = 1:3
subplot(3,1,k)
generatePDF(x,color_scheme_set1(2,:),'hist',no_of_bins(k))
title(sprintf('Bins = %d',no_of_bins(k)))
end
% Loading colors
run colors_definitions.m
% Generating random data
x = randn(1,1e4);
y = 2 + randn(1,1e4);
figure;
generatePDF(x,color_scheme_set1(1,:),'area')
hold on
generatePDF(y,color_scheme_set1(2,:),'area')
legend('Data 1','Data 2'); legend boxoff
% Loading colors
run colors_definitions.m
% Generating random data
x = randn(1,1e4);
% Specifying the number of bins
no_of_bins = 50;
% Data for saving figure in png format (4 inputs required)
savefig = 1; % 1 --> you want to save figure
fig_name = 'generatePDF6';
fig_width_cm = 16;
fig_height_cm = 10;
figure;
generatePDF(x,color_scheme_set1(3,:),'area',no_of_bins,...
savefig,fig_name,fig_width_cm,fig_height_cm)
% Loading colors
run colors_definitions.m
% Generating random data
num_samples = 2e4;
x(1,:) = randn(1,num_samples);
x(2,:) = 2 + 1.25 * randn(1,num_samples);
x(3,:) = 4 + randn(1,num_samples);
x(4,:) = 6 + 0.9 * randn(1,num_samples);
x(5,:) = 8 + 1.5 * randn(1,num_samples);
x(6,:) = 10 + 0.9 * randn(1,num_samples);
x(7,:) = 13 + 1.1 * randn(1,num_samples);
x(8,:) = 16 + 0.9 * randn(1,num_samples);
% Color schemes to test
no_color_schemes = 5;
color_scheme{1} = color_scheme_npg;
color_scheme{2} = color_scheme_aaas;
color_scheme{3} = color_scheme_nejm;
color_scheme{4} = color_scheme_lancet;
color_scheme{5} = color_scheme_set1;
titles_schemes = {'NPG color scheme','AAAS color scheme',...
'NEJM color scheme','LANCET color scheme','Set1 (Brewermap) color scheme'};
figure;
for m = 1:no_color_schemes
subplot(no_color_schemes,1,m)
for k = 1:8
generatePDF(x(k,:),color_scheme{m}(k,:),'area')
hold on
end
title(titles_schemes{m})
legend('Data 1','Data 2','Data 3',...
'Data 4','Data 5','Data 6',...
'Data 7','Data 8','Location','NorthWest'); legend boxoff
end
Generating heatmap in MATLAB using the default function heatmap.m
(introduced in version 2017a) is quite useful for visualizing the magnitude of elements in matrices. However, the size of the generated heatmaps requires a lot of tweaking to produce a reasonable figure. Here, I have written a wrapper code for making nice appropriate-sized heatmaps quickly and efficiently with minimum input.
Function: figure_heatmap.m
% Generating data
x = randn(10,5);
C = corrcoef(x);
% Heatmap figure
figure_heatmap(C);
%Using data of example 1
colorscheme = 'BuGn';
%Requires brewermap package
%Download from https://github.com/DrosteEffect/BrewerMap/blob/master/brewermap.m
% Heatmap figure
figure_heatmap(C,colorscheme);
%Using data of example 1
colorscheme = 'BuGn';
%Requires brewermap package
%Download from https://github.com/DrosteEffect/BrewerMap/blob/master/brewermap.m
text_title = 'Correlation Matrix';
text_labels = {'Variable','Variable'};
% Heatmap figure
figure_heatmap(C,colorscheme,text_title,text_labels);
%Using data of example 1
colorscheme = 'BuGn';
%Requires brewermap package
%Download from https://github.com/DrosteEffect/BrewerMap/blob/master/brewermap.m
text_title = 'Correlation Matrix';
text_labels = {'Variable','Variable'};
limits_data = [-1 1]; %for correlation matrix
text_labels_cells{1} = {'A','B','C','D','E'}; %x-axis cell labels
text_labels_cells{2} = {'A','B','C','D','E'}; %y-axis cell labels
% Heatmap figure
figure_heatmap(C,colorscheme,text_title,text_labels,limits_data,text_labels_cells);
% Generating rectangular matrix data
X = randn(10,6);
colorscheme = 'YlGnBu';
%Requires brewermap package
%Download from https://github.com/DrosteEffect/BrewerMap/blob/master/brewermap.m
text_title = 'Tall Matrix';
text_labels = {'Variable 1','Variable 2'};
limits_data = [floor(min(X(:))) ceil(max(X(:)))];
text_labels_cells{1} = 1:size(X,2); %x-axis cell labels
text_labels_cells{2} = 1:size(X,1); %y-axis cell labels
savefig = 1;
savefig_name = 'heatmap_example5a';
% Heatmap figure
figure_heatmap(X,colorscheme,text_title,text_labels,limits_data,text_labels_cells,...
savefig,savefig_name);
%
Y = randn(11,20);
colorscheme = 'BuPu';
%Requires brewermap package
%Download from https://github.com/DrosteEffect/BrewerMap/blob/master/brewermap.m
text_title = 'Fat Matrix';
text_labels = {'Variable 1','Variable 2'};
limits_data = [floor(min(X(:))) ceil(max(X(:)))];
text_labels_cells{1} = 1:size(Y,2); %x-axis cell labels
text_labels_cells{2} = 1:size(Y,1); %y-axis cell labels
savefig = 1;
savefig_name = 'heatmap_example5b';
% Heatmap figure
figure;
figure_heatmap(Y,colorscheme,text_title,text_labels,limits_data,text_labels_cells,...
savefig,savefig_name);
% Generating rectangular matrix data
X = randn(15,50);
colorscheme = 'BuPu';
%Requires brewermap package
%Download from https://github.com/DrosteEffect/BrewerMap/blob/master/brewermap.m
text_title = 'Big Rectangular Matrix';
text_labels = {'Variable 1','Variable 2'};
limits_data = [floor(min(X(:))) ceil(max(X(:)))];
text_labels_cells{1} = 1:size(X,2); %x-axis cell labels
text_labels_cells{2} = 1:size(X,1); %y-axis cell labels
savefig = 1;
savefig_name = 'heatmap_example6';
% Heatmap figure
figure_heatmap(X,colorscheme,text_title,text_labels,limits_data,text_labels_cells,...
savefig,savefig_name);
In progress ...
If you find this repo useful, please cite it using the following information:
Ahmed Abdul Quadeer. (2019, December 18). ahmedaq/Making-elegant-Matlab-figures: Release v1.0 (Version v1.0). Zenodo. http://doi.org/10.5281/zenodo.3582848
@software{ahmed_abdul_quadeer_2019_3582848, author = {Ahmed Abdul Quadeer}, title = {{ahmedaq/Making-elegant-Matlab-figures: Release v1.0}}, month = dec, year = 2019, publisher = {Zenodo}, version = {v1.0}, doi = {10.5281/zenodo.3582848}, url = {https://doi.org/10.5281/zenodo.3582848} }