delete zeros rows and columns
405 views (last 30 days)
Show older comments
Andrea
on 31 May 2012
Commented: Walter Roberson
on 14 Jan 2025
I try this code to delete all row and column with all zero values in them. It simply works for deleting the columns with all zero values abut it does not work for rows! Can anybody please help me?
data=[0 0 0 0 0 0 0 0; 0 0 2 3 4 0 1 0; 0 0 1 2 3 0 0 0];
data( all( ~any( data), 2 ), : ) = []; % removes all rows with all zero
data( :, all( ~any( data ), 1 ) ) = []; % and columns
I mean the first line (% removes all rows with all zero) does not work!
0 Comments
Accepted Answer
Walter Roberson
on 31 May 2012
data( ~any(data,2), : ) = []; %rows
data( :, ~any(data,1) ) = []; %columns
14 Comments
Gabriel
on 14 Jan 2025
Dears,
while running this matlab code, the transpose of the matrix teste s not working. Can u please suggest me if there is updated code for running a contourf for R2024b. [Error in (line 145).... contourf(linspace(0,3,42),linspace(0.0,0.1,40),teste','LineStyle','-','LineWidth',0.2)]
contourf(linspace(0,3,42),linspace(0.0,0.1,40),teste','LineStyle','-','LineWidth',0.2)
%%
when I define the teste matrix it is in line 140, teste = zeros(42,40);
I got anather problem in in line 160, mc6 = set(m1,'parameters',c6);
%%below I post the array sizes, the connection between the for loop and the contourf(). could you check if the matrix teste is correctly defined. and suggest me where the problem to fix. Thanks
%%%
clear
close all
clc
%%
% Choose shock variance. If solution order = 1, high_var is ok.
label_1={1,'low_var'
2,'high_var'};
% Choose exogenous switching, endogenous switching or fixed regimes.
label_2={1,'fixed_regime_M'
2,'ms_regimes'
3,'ms_endog_regimes'
4,'fixed_regime_F'};
% Choose taxation scheme
label_3={1,'non-distortionary'
2,'distortionary'};
% Read the models and their calibrations. Change labels as disired!
m0=rise('rsa_ms',...
'rise_flags',...
struct('varmodel',label_1{2,1},...
'msmodel', label_2{3,1},...
'taxdistortionary', label_3{1,1}),...
'irf_anticipate',false);
Unrecognized function or variable 'rise'.
% Solve the model with baseline calibration; 1st order. Solvers: 'mfi', 'mn', 'mnk', 'fwz'. See issue 108.
% "The newton algorithm mn is locally stable around an equilibrium while the main strength of mfi is to handle large systems."
m1=solve(m0,'solve_shock_horizon',0,'solve_order',1,...
'solve_check_stability',false); %'solver','mn','solve_perturbation_type','mw' or 'frwz'
%[malt,stab]=solve_alternatives(m0);
% 2ns order solution
%m2nd=solve(m1,'solve_order',2,'solve_derivatives_type','automatic');
%m2nd.print_solution({'Pi','Y','C','b','R','sp','Welfare','CE'});
% Check the stability of the system
%m1.is_stable_system % Stability is denoted by 1 (0 means unstable)
% Print solution and steady state values
% m1.print_solution() % for all variables
% print solution for a subset of variables only
m1.print_solution({'Pi','Y','C','B','R','sp','Welfare','CE','Q'});
%% QUICK IRFs
myirfs=irf(m1,'irf_periods',24,'irf_shock_sign',1);
var_list={'log_y','C','pi_ann','B_nominal','B','sp','i_ann','r_real_ann','Welfare'};
figure('name','Impulse responses to a fiscal shock');
for ii=1:numel(var_list)
subplot(3,3,ii)
reg1=myirfs.eps_G.(var_list{ii}).data(:,1);
reg2=myirfs.eps_G.(var_list{ii}).data(:,2);
plot(reg1);
hold on
plot(reg2);
title(var_list{ii})
axis tight
end
%% STABILITY & SOLUTIONS w.r.t. BASELINE CALIBRATION
% SELECTED PARAMETERS: t_ss, delta_B, abc_tp_1_2, rho_G, rho_T, kappa.
grid_size = 20;
t_ss_vec = linspace(0,0.03,grid_size);
delta_B_vec = linspace(0,1.00,grid_size);
rho_G_vec = linspace(0,0.90,grid_size);
rho_T_vec = linspace(0,0.90,grid_size);
kappa_vec = linspace(0,0.98,grid_size);
prob_vec = linspace(0,0.20,grid_size);
solution_mat = NaN(grid_size,6);
stab_mat = NaN(grid_size,6);
%% Loop for just one parameter
for j = 1:grid_size
c = struct('delta_B_abc_1', delta_B_vec(j)); % CHANGE THIS LINE FOR DESIRED PARAMETER
mc = set(m1,'parameters',c);
ms=solve(mc,'solve_order',1,'solve_check_stability',false,'solve_perturbation_type','mw');
solution_mat(j,1) = ms.nsols;
disp(j)
end
%% Loop for two parameter
grid_size = 20;
delta_B_vec = linspace(0, 0.05, grid_size);
phi_pi_vec = linspace(0, 3.00, grid_size+1);
for j = 1:(grid_size+1)
for i = 1:grid_size
c = struct('delta_B_abc_1', delta_B_vec(i),'phi_pi_abc_1', phi_pi_vec(j)); % _abc_1
mc = set(m0,'parameters',c);
ms=solve(mc,'solve_order',1);
if ms.nsols == 0
% d_mat(j,i) = NaN;
% pi_mat(j,i) = NaN;
% var_mat(j,i) = NaN;
solution_mat(j,i) = 0;
stab_mat(j,i) = 0;
else
% simdata0 = simulate(ms,'simul_regime',1,'simul_order',1,'simul_periods',1000);
% if isempty(simdata0)
% d_mat(j,i) = NaN;
% pi_mat(j,i) = NaN;
% stab_mat(j,i) = 0;
% solution_mat(j,i) = 0;
% var_mat(j,i) = NaN;
% else
% tempirf=irf(ms,'irf_periods',24,'irf_shock_sign',1);
% d_mat(j,i) = (tempirf.eps_G.B.data(4,1) - tempirf.eps_G.B.data(13,1)) > 0;
% pi_mat(j,i) = mean(simdata0.Pi.data);
% var_mat(j,i) = var(simdata0.Pi.data);
solution_mat(j,i) = ms.nsols;
stab_mat(j,i) = ms.is_stable_system;
% end
end
end
disp((j))
end
%figure(1)
%s = pcolor(stab_mat');
%s.LineStyle = '-';
%colormap(gray)
%teste = zeros(42,40);
%% Plot stability
figure(1)
set(1,'Color','w')
contourf(linspace(0,3,42),linspace(0.0,0.1,40),teste','LineStyle','-','LineWidth',0.2)
colormap(gray)
set(gca,'fontsize',10)
xlabel('Taylor Rule Parameter - $\phi_\pi$','interpreter','latex')
ylabel('Fiscal Parameter - $\delta_B$','interpreter','latex')
set(gca,'TickLabelInterpreter','latex')
%yticklabels([0.00 0.02 0.04 0.06 0.08 1.00])
set(gca,'fontsize',12)
line1 = '';
%line2 = 'Fixed Parameters';
%line2 = 'Exogenous Switching';
line2 = 'Endogenous Switching';
title(sprintf('\\begin{tabular}{c} %s %s %s \\end{tabular}',line1,'\\',line2),...
'interpreter','latex')
x0=1000;
y0=300;
width=350;
height=300;
set(1,'position',[x0,y0,width,height]);
%% Loop all selected parameters
for j = 1:grid_size
% 'c' stands for 'calibration'
c1 = struct('t_ss', t_ss_vec(j));
c2 = struct('delta_B_abc_1', delta_B_vec(j));
c3 = struct('rho_G', rho_G_vec(j));
c4 = struct('rho_T', rho_T_vec(j));
c5 = struct('kappa', kappa_vec(j));
c6 = struct('abc_tp_1_2', prob_vec(j));
% 'mc' stands for 'model with new calibration'
mc1 = set(m1,'parameters',c1);
mc2 = set(m1,'parameters',c2);
mc3 = set(m1,'parameters',c3);
mc4 = set(m1,'parameters',c4);
mc5 = set(m1,'parameters',c5);
mc6 = set(m1,'parameters',c6);
% Solve
ms1=solve(mc1,'solve_order',1,'solve_check_stability',false);
ms2=solve(mc2,'solve_order',1,'solve_check_stability',false);
ms3=solve(mc3,'solve_order',1,'solve_check_stability',false);
ms4=solve(mc4,'solve_order',1,'solve_check_stability',false);
ms5=solve(mc5,'solve_order',1,'solve_check_stability',false);
ms6=solve(mc6,'solve_order',1,'solve_check_stability',false);
% Has a solution? Is stable?
solution_mat(j,1) = ms1.nsols;
solution_mat(j,2) = ms2.nsols;
solution_mat(j,3) = ms3.nsols;
solution_mat(j,4) = ms4.nsols;
solution_mat(j,5) = ms5.nsols;
solution_mat(j,6) = ms6.nsols;
disp(j)
end
More Answers (2)
Geoff
on 31 May 2012
Yeah you did too much!
% Remove zero rows
data( all(~data,2), : ) = [];
% Remove zero columns
data( :, all(~data,1) ) = [];
5 Comments
Rubina Easmin
on 10 Feb 2020
I'm new to MATLAB, I want to delete the entire rows and columns which contain all of ones in a binary image.I want to keep rows and columns if that contain only single zero. Any tips on how to go about doing this? Thanks
>> binaryimage
binaryimage =
33×35 logical array
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 1
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 1 1 1 1 1 1 1 1 0 0 1
1 1 1 1 1 1 0 1 1 1 1 1 1 1 0 0 0 0 0 0 0 1 1 1 1 1 1 0 0 1
1 1 1 1 1 1 0 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 1 1 1 1 1 0 0 1
1 1 1 1 1 1 0 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 1 1 1 1 0 0 1
1 1 1 1 1 1 0 0 1 1 1 1 1 1 0 0 0 0 0 0 1 1 0 0 1 1 1 0 0 1
1 1 1 1 1 1 1 0 1 1 1 1 1 1 1 1 0 0 1 1 1 1 0 0 1 1 1 0 0 1
1 1 1 1 1 1 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 1 1 1 0 0 1
1 1 1 1 1 1 1 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 1 1 0 0 1
1 1 1 1 1 1 1 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 1 1 0 0 1
1 1 1 1 1 1 1 1 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 0 1 1 1 0 0 1
1 1 1 1 1 1 1 1 1 0 0 1 1 1 1 1 1 1 1 1 1 1 1 0 1 1 1 0 0 1
1 1 1 1 1 1 1 1 1 1 0 0 1 1 1 1 1 1 1 1 1 1 1 0 0 1 1 0 0 1
1 1 1 1 1 1 1 1 1 1 0 0 0 1 1 1 1 1 1 1 1 1 0 0 0 0 1 0 0 1
1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 1 1 1 1 1 1 0 0 1 0 0 1 0 0 1
1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 1 1 0 0 0 0 0 1
1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 1 1 1 1 0 0 0 0 1
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 1 1 1 1 1 1 1 0 0 0 1
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 1
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 1
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 1
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 1
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 1
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
I want to remove rows and columns that contains all of ones
6 Comments
Gabriel
on 14 Jan 2025
Edited: Walter Roberson
on 14 Jan 2025
Dears,
while running this matlab code, the transpose of the matrix teste s not working. Can u please suggest me if there is updated code for running a contourf for R2024b. [Error in (line 145).... contourf(linspace(0,3,42),linspace(0.0,0.1,40),teste','LineStyle','-','LineWidth',0.2)]
contourf(linspace(0,3,42),linspace(0.0,0.1,40),teste','LineStyle','-','LineWidth',0.2)
%%
when I define the teste matrix it is in line 140, teste = zeros(42,40);
I got anather problem in in line 160, mc6 = set(m1,'parameters',c6);
%%below I post the array sizes, the connection between the for loop and the contourf(). could you check if the matrix teste is correctly defined. and suggest me where the problem to fix. Thanks
%%%
clear
close all
clc
%%
% Choose shock variance. If solution order = 1, high_var is ok.
label_1={1,'low_var'
2,'high_var'};
% Choose exogenous switching, endogenous switching or fixed regimes.
label_2={1,'fixed_regime_M'
2,'ms_regimes'
3,'ms_endog_regimes'
4,'fixed_regime_F'};
% Choose taxation scheme
label_3={1,'non-distortionary'
2,'distortionary'};
% Read the models and their calibrations. Change labels as disired!
m0=rise('rsa_ms',...
'rise_flags',...
struct('varmodel',label_1{2,1},...
'msmodel', label_2{3,1},...
'taxdistortionary', label_3{1,1}),...
'irf_anticipate',false);
Unrecognized function or variable 'rise'.
% Solve the model with baseline calibration; 1st order. Solvers: 'mfi', 'mn', 'mnk', 'fwz'. See issue 108.
% "The newton algorithm mn is locally stable around an equilibrium while the main strength of mfi is to handle large systems."
m1=solve(m0,'solve_shock_horizon',0,'solve_order',1,...
'solve_check_stability',false); %'solver','mn','solve_perturbation_type','mw' or 'frwz'
%[malt,stab]=solve_alternatives(m0);
% 2ns order solution
%m2nd=solve(m1,'solve_order',2,'solve_derivatives_type','automatic');
%m2nd.print_solution({'Pi','Y','C','b','R','sp','Welfare','CE'});
% Check the stability of the system
%m1.is_stable_system % Stability is denoted by 1 (0 means unstable)
% Print solution and steady state values
% m1.print_solution() % for all variables
% print solution for a subset of variables only
m1.print_solution({'Pi','Y','C','B','R','sp','Welfare','CE','Q'});
%% QUICK IRFs
myirfs=irf(m1,'irf_periods',24,'irf_shock_sign',1);
var_list={'log_y','C','pi_ann','B_nominal','B','sp','i_ann','r_real_ann','Welfare'};
figure('name','Impulse responses to a fiscal shock');
for ii=1:numel(var_list)
subplot(3,3,ii)
reg1=myirfs.eps_G.(var_list{ii}).data(:,1);
reg2=myirfs.eps_G.(var_list{ii}).data(:,2);
plot(reg1);
hold on
plot(reg2);
title(var_list{ii})
axis tight
end
%% STABILITY & SOLUTIONS w.r.t. BASELINE CALIBRATION
% SELECTED PARAMETERS: t_ss, delta_B, abc_tp_1_2, rho_G, rho_T, kappa.
grid_size = 20;
t_ss_vec = linspace(0,0.03,grid_size);
delta_B_vec = linspace(0,1.00,grid_size);
rho_G_vec = linspace(0,0.90,grid_size);
rho_T_vec = linspace(0,0.90,grid_size);
kappa_vec = linspace(0,0.98,grid_size);
prob_vec = linspace(0,0.20,grid_size);
solution_mat = NaN(grid_size,6);
stab_mat = NaN(grid_size,6);
%% Loop for just one parameter
for j = 1:grid_size
c = struct('delta_B_abc_1', delta_B_vec(j)); % CHANGE THIS LINE FOR DESIRED PARAMETER
mc = set(m1,'parameters',c);
ms=solve(mc,'solve_order',1,'solve_check_stability',false,'solve_perturbation_type','mw');
solution_mat(j,1) = ms.nsols;
disp(j)
end
%% Loop for two parameter
grid_size = 20;
delta_B_vec = linspace(0, 0.05, grid_size);
phi_pi_vec = linspace(0, 3.00, grid_size+1);
for j = 1:(grid_size+1)
for i = 1:grid_size
c = struct('delta_B_abc_1', delta_B_vec(i),'phi_pi_abc_1', phi_pi_vec(j)); % _abc_1
mc = set(m0,'parameters',c);
ms=solve(mc,'solve_order',1);
if ms.nsols == 0
% d_mat(j,i) = NaN;
% pi_mat(j,i) = NaN;
% var_mat(j,i) = NaN;
solution_mat(j,i) = 0;
stab_mat(j,i) = 0;
else
% simdata0 = simulate(ms,'simul_regime',1,'simul_order',1,'simul_periods',1000);
% if isempty(simdata0)
% d_mat(j,i) = NaN;
% pi_mat(j,i) = NaN;
% stab_mat(j,i) = 0;
% solution_mat(j,i) = 0;
% var_mat(j,i) = NaN;
% else
% tempirf=irf(ms,'irf_periods',24,'irf_shock_sign',1);
% d_mat(j,i) = (tempirf.eps_G.B.data(4,1) - tempirf.eps_G.B.data(13,1)) > 0;
% pi_mat(j,i) = mean(simdata0.Pi.data);
% var_mat(j,i) = var(simdata0.Pi.data);
solution_mat(j,i) = ms.nsols;
stab_mat(j,i) = ms.is_stable_system;
% end
end
end
disp((j))
end
%figure(1)
%s = pcolor(stab_mat');
%s.LineStyle = '-';
%colormap(gray)
%teste = zeros(42,40);
%% Plot stability
figure(1)
set(1,'Color','w')
contourf(linspace(0,3,42),linspace(0.0,0.1,40),teste','LineStyle','-','LineWidth',0.2)
colormap(gray)
set(gca,'fontsize',10)
xlabel('Taylor Rule Parameter - $\phi_\pi$','interpreter','latex')
ylabel('Fiscal Parameter - $\delta_B$','interpreter','latex')
set(gca,'TickLabelInterpreter','latex')
%yticklabels([0.00 0.02 0.04 0.06 0.08 1.00])
set(gca,'fontsize',12)
line1 = '';
%line2 = 'Fixed Parameters';
%line2 = 'Exogenous Switching';
line2 = 'Endogenous Switching';
title(sprintf('\\begin{tabular}{c} %s %s %s \\end{tabular}',line1,'\\',line2),...
'interpreter','latex')
x0=1000;
y0=300;
width=350;
height=300;
set(1,'position',[x0,y0,width,height]);
%% Loop all selected parameters
for j = 1:grid_size
% 'c' stands for 'calibration'
c1 = struct('t_ss', t_ss_vec(j));
c2 = struct('delta_B_abc_1', delta_B_vec(j));
c3 = struct('rho_G', rho_G_vec(j));
c4 = struct('rho_T', rho_T_vec(j));
c5 = struct('kappa', kappa_vec(j));
c6 = struct('abc_tp_1_2', prob_vec(j));
% 'mc' stands for 'model with new calibration'
mc1 = set(m1,'parameters',c1);
mc2 = set(m1,'parameters',c2);
mc3 = set(m1,'parameters',c3);
mc4 = set(m1,'parameters',c4);
mc5 = set(m1,'parameters',c5);
mc6 = set(m1,'parameters',c6);
% Solve
ms1=solve(mc1,'solve_order',1,'solve_check_stability',false);
ms2=solve(mc2,'solve_order',1,'solve_check_stability',false);
ms3=solve(mc3,'solve_order',1,'solve_check_stability',false);
ms4=solve(mc4,'solve_order',1,'solve_check_stability',false);
ms5=solve(mc5,'solve_order',1,'solve_check_stability',false);
ms6=solve(mc6,'solve_order',1,'solve_check_stability',false);
% Has a solution? Is stable?
solution_mat(j,1) = ms1.nsols;
solution_mat(j,2) = ms2.nsols;
solution_mat(j,3) = ms3.nsols;
solution_mat(j,4) = ms4.nsols;
solution_mat(j,5) = ms5.nsols;
solution_mat(j,6) = ms6.nsols;
disp(j)
end
See Also
Categories
Find more on Designs in Help Center and File Exchange
Products
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!