I have the following gabor filter to extract image texture feature..
a=imread('image0001.jpg');
a=double(a);
a=a-mean(a(:));
[r,c,l]=size(a);
K=5; S=6;
Uh=0.4;
Ul=0.05;
alpha=(Uh/Ul)^(1/(S-1));
sigmau=(alpha-1)*Uh/((alpha+1)*sqrt(2*log(2)));
sigmav=tan(pi/(2*K))*(Uh-2*log(2)*((sigmau^2)/Uh))/sqrt((2*log(2))-(((2*log(2))^2)*(sigmau^2)/(Uh^2)));
sigmax=1/(2*pi*sigmau);
sigmay=1/(2*pi*sigmav);
b=fft2(a);
[e d]=size(b);
i=1;
G=zeros(r,c,S*K);
IZ=zeros(r,c,S*K);
for m=0:S-1
for n=0:K-1
fprintf(1,'.');
for x=-r/2+1:r/2;
for y=-c/2+1:c/2;
xdash=(alpha^(-m))*((x)*cos(n*pi/K)+(y)*sin(n*pi/K));
ydash=(alpha^(-m))*((y)*cos(n*pi/K)-(x)*sin(n*pi/K));
g(r/2+x,r/2+y)=(alpha^(-m))*((1/(2*pi*sigmax*sigmay))*exp(-0.5*(((xdash^2)/(sigmax^2))+((ydash^2)/(sigmay^2)))+0.8i*pi*xdash));
end
end
[rr cc]=size(g);
G(:,:,i)=g;
h=fft2(g);
z=b.*h;
iz=ifft2(z);
IZ(:,:,i)=iz;
FeatureVector(i)=mean(abs(iz(:)));
i=i+1;
end
end
fprintf(1,'\n');
%%%%%%%%%
When I run this code I get this Error:
Error using ==> times Matrix
dimensions must agree. Error in ==>
ComputeGaborFeatures4 at 37
z=b.*h;
Please if any one can help me to solve this error or any one can give me another simple gabor filter?
The error is due to the calling of Array Multiplication (.*) with b and h of non equal size, because rr doesn't equal r and cc doesn't c.
Either you wanted to use Matrix Multiplication (*) or you need to make g and a the same size before calling fft2.
the error might change the g(r/2+x,r/2+y) to g(r/2+x,c/2+y), the the
Related
I have read a lot about the pain of replicate the easy robust option from STATA to R to use robust standard errors. I replicated following approaches: StackExchange and Economic Theory Blog. They work but the problem I face is, if I want to print my results using the stargazer function (this prints the .tex code for Latex files).
Here is the illustration to my problem:
reg1 <-lm(rev~id + source + listed + country , data=data2_rev)
stargazer(reg1)
This prints the R output as .tex code (non-robust SE) If i want to use robust SE, i can do it with the sandwich package as follow:
vcov <- vcovHC(reg1, "HC1")
if I now use stargazer(vcov) only the output of the vcovHC function is printed and not the regression output itself.
With the package lmtest() it is possible to print at least the estimator, but not the observations, R2, adj. R2, Residual, Residual St.Error and the F-Statistics.
lmtest::coeftest(reg1, vcov. = sandwich::vcovHC(reg1, type = 'HC1'))
This gives the following output:
t test of coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -2.54923 6.85521 -0.3719 0.710611
id 0.39634 0.12376 3.2026 0.001722 **
source 1.48164 4.20183 0.3526 0.724960
country -4.00398 4.00256 -1.0004 0.319041
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
How can I add or get an output with the following parameters as well?
Residual standard error: 17.43 on 127 degrees of freedom
Multiple R-squared: 0.09676, Adjusted R-squared: 0.07543
F-statistic: 4.535 on 3 and 127 DF, p-value: 0.00469
Did anybody face the same problem and can help me out?
How can I use robust standard errors in the lm function and apply the stargazer function?
You already calculated robust standard errors, and there's an easy way to include it in the stargazeroutput:
library("sandwich")
library("plm")
library("stargazer")
data("Produc", package = "plm")
# Regression
model <- plm(log(gsp) ~ log(pcap) + log(pc) + log(emp) + unemp,
data = Produc,
index = c("state","year"),
method="pooling")
# Adjust standard errors
cov1 <- vcovHC(model, type = "HC1")
robust_se <- sqrt(diag(cov1))
# Stargazer output (with and without RSE)
stargazer(model, model, type = "text",
se = list(NULL, robust_se))
Solution found here: https://www.jakeruss.com/cheatsheets/stargazer/#robust-standard-errors-replicating-statas-robust-option
Update I'm not so much into F-Tests. People are discussing those issues, e.g. https://stats.stackexchange.com/questions/93787/f-test-formula-under-robust-standard-error
When you follow http://www3.grips.ac.jp/~yamanota/Lecture_Note_9_Heteroskedasticity
"A heteroskedasticity-robust t statistic can be obtained by dividing an OSL estimator by its robust standard error (for zero null hypotheses). The usual F-statistic, however, is invalid. Instead, we need to use the heteroskedasticity-robust Wald statistic."
and use a Wald statistic here?
This is a fairly simple solution using coeftest:
reg1 <-lm(rev~id + source + listed + country , data=data2_rev)
cl_robust <- coeftest(reg1, vcov = vcovCL, type = "HC1", cluster = ~
country)
se_robust <- cl_robust[, 2]
stargazer(reg1, reg1, cl_robust, se = list(NULL, se_robust, NULL))
Note that I only included cl_robust in the output as a verification that the results are identical.
I have an error for the following code in Julia for solving NLP problem.
using JuMP
using Ipopt
using DataFrames,ConditionalJuMP
m = Model(solver=IpoptSolver())
#importing data
xi=[-1.016713795 0.804447648 0.932137661 1.064136698 -0.963217531
-1.048396778 1.076371484 1.099027177 1.061556926 -0.95185481
-0.980261302 0.271253807 0.184946729 1.062838197 -0.958794505
-0.980703191 0.278820569 0.231132677 1.062967459 -0.959302488
-0.953074503 -0.00768112 0.128808175 1.067743779 -0.978524489
-1.014866259 0.815325963 1.065956208 1.067059122 -0.974682291
-0.995088119 0.550359991 0.845087207 1.066556784 -0.973154887]
xj=xi
pii=[-300
-259.6530828
-284.3708955
-291.3387625
-342.4479859
-356.5031603
-351.0154738]
sample_size=7
bus_num=5
tempsum=0
#define variables
#variable(m,aij[1:2*bus_num,1:2*bus_num]) # define matrix by
2bus_num*2bus_num
for n=1:sample_size
for i=1:bus_num
for j=1:bus_num
tempsum=(aij[i,j]*xi[n,i]*xj[n,j]-pii[n,2])^2+tempsum
end
end
end
#define constraints
#constraint(m, [i=1:2*bus_num,j=2*bus_num],aij[i,j]==aij[j,i])
#constraint(m, [i=1:2*bus_num,j=2*bus_num],aij[i,j]*aij[i,j]<=aij
[i,i]*aij[j,j])
#constraint(m, [i=1:2*bus_num,j=2*bus_num],aij[i,i]*aij[j,j]<=0.25*
(aij[i,i]+aij[j,j])*(aij[i,i]+aij[j,j]))
#define NLP objective function
#NLobjective(m, Min, tempsum)
solve(m)
println("m = ",getobjectivevalue(m))
println("Aij = ", getvalue(aij))
but after operation, an error was shown as below.
MethodError: no method matching parseNLExpr_runtime(::JuMP.Model,
::JuMP.GenericQuadExpr{Float64,JuMP.Variable},
::Array{ReverseDiffSparse.NodeData,1}, ::Int64, ::Array{Float64,1})
The error occurs on #NLobjective(m, Min, tempsum), but I dont know how to modify the code. May you please anyone help me ?
Besides, I am also confused by how to add positive definite contraints here for the solution Matrix aij, as I want to get a positive definite matrix aij.
#NLobjective(m, Min, tempsum) the tempsum in this code must be a function which is registered by Jump.register(***).
I found the proper syntax to import my centriod factor extraction into SPSS and rotate it. (The semi-bumbling tale is here.
The next issue is this: because of limitations in SPSS on what subcommands can be used when reading a matrix in (only /ROTATE and /EXTRACTION are permitted), I can't get the factor scores. SPSS displays this error:
"Factor scores cannot be computed with matrix input."
I still need to find a way to get the FSCORE of the newly rotated factors of each factor for all cases by running a regression using the newly rotated factors and saving the regression values as a new variable (/SAVE REG(ALL).
Ideas are welcome. Thank you for your expertise!
Assets: Dataset A of 36 cases and 74 variables (the basis of the centriod factor extraction); centriod factor extraction matrix
Here's the SPSS syntax that almost does what I need - except it uses PCA extraction instead of centroid.
FACTOR
/VARIABLES VAR00001 VAR00002 VAR00003 VAR00004 VAR00005 VAR00006 VAR00007 VAR00008 VAR00009 VAR00010 VAR00011 VAR00012 VAR00013 VAR00014 VAR00015 VAR00016 VAR00017 VAR00018 VAR00019 VAR00020 VAR00021 VAR00022 VAR00023 VAR00024 VAR00025 VAR00026 VAR00027 VAR00028 VAR00029 VAR00030 VAR00031 VAR00032 VAR00033 VAR00034 VAR00035 VAR00036 VAR00037 VAR00038 VAR00039 VAR00040 VAR00041 VAR00042 VAR00043 VAR00044 VAR00045 VAR00046 VAR00047 VAR00048 VAR00049 VAR00050 VAR00051 VAR00052 VAR00053 VAR00054 VAR00055 VAR00056 VAR00057 VAR00058 VAR00059 VAR00060 VAR00061 VAR00062 VAR00063 VAR00064 VAR00065 VAR00066 VAR00067 VAR00068 VAR00069 VAR00070 VAR00071 VAR00072 VAR00073 VAR00074
/MISSING LISTWISE
/ANALYSIS VAR00001 VAR00002 VAR00003 VAR00004 VAR00005 VAR00006 VAR00007 VAR00008 VAR00009 VAR00010 VAR00011 VAR00012 VAR00013 VAR00014 VAR00015 VAR00016 VAR00017 VAR00018 VAR00019 VAR00020 VAR00021 VAR00022 VAR00023 VAR00024 VAR00025 VAR00026 VAR00027 VAR00028 VAR00029 VAR00030 VAR00031 VAR00032 VAR00033 VAR00034 VAR00035 VAR00036 VAR00037 VAR00038 VAR00039 VAR00040 VAR00041 VAR00042 VAR00043 VAR00044 VAR00045 VAR00046 VAR00047 VAR00048 VAR00049 VAR00050 VAR00051 VAR00052 VAR00053 VAR00054 VAR00055 VAR00056 VAR00057 VAR00058 VAR00059 VAR00060 VAR00061 VAR00062 VAR00063 VAR00064 VAR00065 VAR00066 VAR00067 VAR00068 VAR00069 VAR00070 VAR00071 VAR00072 VAR00073 VAR00074
/PRINT INITIAL CORRELATION SIG DET INV REPR AIC EXTRACTION ROTATION FSCORE
/FORMAT BLANK(.544)
/CRITERIA FACTORS(6) ITERATE(80)
/EXTRACTION PC <---Here's the rub.
/CRITERIA ITERATE(80) DELTA(0)
/ROTATION OBLIMIN
/SAVE REG(ALL)
/METHOD=CORRELATION.
I referred to the MATRIX command in my reply to your other post (Rotations). You will need to research the appropriate equations for performing this calculation and set up the matrix algebra within a MATRIX END - MATRIX code block. Easy once you have the math right. I'm too busy/lazy to research and write it but this should provide a fertile lead.
I am currently runnuing training in matlab on a matrix of logspecrum samples I am constantly dealing with underflow problems.I understood that I need to work with log's in order to deal with underflowing.
I am still strugling with uderflow though , when i calculate the mean (mue) bucause it is negetive i cant work with logs so i need the real values that underflow.
These are equasions i am working with:
In MATLAB code i calulate log_tau in oreder avoid underflow but when calulating mue i need exp(log(tau)) which goes to zero.
I am attaching relevent MATLAB code
**in the code i called the variable alpha is tau ...
for i = 1 : 50
log_c = Logsum(log_alpha,1) - log(N);
c = exp(log_c);
mue = DataMat*alpha./(repmat(exp(Logsum(log_alpha,1)),FrameSize,1));
log_abs_mue = log(abs(mue));
log_SigmaSqr = log((DataMat.^2)*alpha) - repmat(Logsum(log_alpha,1),FrameSize,1) - 2*log_abs_mue;
SigmaSqr = exp(log_SigmaSqr);
for j=1:N
rep_DataMat(:,:,j) = repmat(DataMat(:,j),1,M);
log_gamma(j,:) = log_c - 0.5*(FrameSize*log(2*pi)+sum(log_SigmaSqr)) + sum((rep_DataMat(:,:,j) - mue).^2./(2*SigmaSqr));
end
log_alpha = log_gamma - repmat(Logsum(log_gamma,2),1,M);
alpha = exp(log_alpha);
end
c = exp(log_c);
SigmaSqr = exp(log_SigmaSqr);
does any one see how i can avoid this? or what needs to be fixed in code?
What i did was add this line to the MATLAB code:
mue(isnan(mue))=0; %fix 0/0 problem
and this one:
SigmaSqr(SigmaSqr==0)=1;%fix if mue_k = x_k
not sure if this is the best solution but is seems to work...
any have a better idea?
I am new to this forum. Let me get started: I work on MATLAB and keep getting errors all the time. Finally i found a good forum like yours. My problem is this: I have an image which I want to put inside a large matrix. Everytime I do it I get
??? ERROR: subscripted assignment dimension mismatch
I tried everything possible, like u say resize, repmat, reshape....but I could not guess what is going wrong.
My code is like this:
nem(:,:,1) = image %// <-- error subscripted assignment dimension mismatch
my size of image is
71 * 71
bytes :----40328
class :----double
nem is created by
nem = zeros([size(inputimage,1),size(inputimage,2),12]);
size of inputmage is
[m,n,o] = size(inputimage);
m = 584 n = 565 o = 1
and size of nem:
[m,n,o] = size(img_out);
m = 584 n = 565 o = 12
You are trying to "fit" image a 71-by-71 matrix into mem(:,:,1) which is 584-by-565 matrix.
How do you expect Matlab to do this type of assignment??
You can fit image into a part of mem
>> mem( 1:size(image,1), 1:size(image,2), 1 ) = image