I want to run a video in matlab but I got some problems while doing so. I tried to run the video while using 'load' and then 'implay'. But I got this problem. My movie file is a textfile which constains mostly of zeros, arranged diagonally.
When I load this file in matlab I get different values. Here a snippet:
0 0 0 0 0 0 0 0 0 0 0
0 0 -5 0 0 -5.00097609500000 0 0 0 0 0
0 0 -10 0 0 -10.0019423800000 0 0 0 0 0
0 0 -15 0 0 -15.0028988550000 0 0 0 0 0
0 0 -20 0 0 -20.0038455200000 0 0 0 0 0
0 0 -25 0 0 -25.0047823750000 0 0 0 0 0
0 0 -30 0 0 -30.0057094200000 0 0 0 0 0
0 0 -35 0 0 -35.0066266550000 0 0 0 0 0
0 0 -40 0 0 -40.0075340800000 0 0 0 0 0
0 0 -45 0 0 -45.0084316950000 0 0 0 0 0
I dont know why I get a mat.file with different values and why my movie does not run. I wanted to upload the original textfile but this not possible.
When I try to run I receive following message: There is not enough memory to open 'filename' in the Editor.
I hope you guys can help me!
Related
This function is confusing to use, and it always gives me an error:
To RESHAPE the number of elements must not change.
That's my code:
im=im2col(zeros(300,300),[3 3]);
im(:,9)=ones(9,1);
im=col2im(im,[3 3],[300 300]);
Basically, this code just gets the block at index 6, replaces it with ones, and reassembles it back into the original image. What's wrong with this code?
It seems you want to create distinct blocks from your input array, change single blocks, and rearrange them. (Your target size is the same as your input array size.) So, you must use the distinct parameter in both, im2col as well as col2im:
blk_size = [3, 3];
im = zeros(9, 9)
temp = im2col(im, blk_size, 'distinct');
temp(:, 3) = ones(prod(blk_size), 1);
im2 = col2im(temp, [3 3], size(im), 'distinct')
Output:
im =
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
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
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
im2 =
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
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 0 0 0 0 0 0
1 1 1 0 0 0 0 0 0
1 1 1 0 0 0 0 0 0
When using im2col with the sliding parameter, which is also the default, if no parameter is set at all, there'll be a lot more columns in the result than can be rearranged to the input array size, cf. the Tips section on im2col.
Hope that helps!
I have a folder with pictures. For each picture, I want to take the maximum value and add it to a new matrix (I created a zeros-matrix, so the zeroes will be replaced with the new values).
This is my code:
function handles = original(hObject, eventdata, handles)
handles.weed=handles.selected;
pic=imread(handles.me);
handles.pic=pic;
axes(handles.axes1)
imshow(pic);
num=max(pic(:))
zeroMat = zeros(1,70);
handles.zeroMat = zeroMat;
for i =1:3
if zeroMat(1,i)~= 0;
i=i+1
else
zeroMat(1,i)=num
break
end
end
zeroMat(1,i)=num
Every time I select a new picture, the zeromat restarts itself to a new zeros-matrix. I know why it happens, but unfortunately I don't know how to overcome it.
This is the output:
pic1:
zeroMat =
Columns 1 through 20
255 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
Columns 21 through 40
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
Columns 41 through 60
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
Columns 61 through 70
0 0 0 0 0 0 0 0 0 0
pic2:
Columns 1 through 20
203 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
Columns 21 through 40
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
Columns 41 through 60
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
Columns 61 through 70
0 0 0 0 0 0 0 0 0 0
I can't tell how this function is being invoked, so can't advice how to change up the logic. A rough fix could be to include the line
if ~isfield(handles,'zeroMat')
handles.zeroMat = zeros(1,70);
end % if
which should create handles.zeroMat the first time the function is run. You could then do something like
firstNonzero = find(handles.zeroMat > 0, 1, 'first'); % 'first' not needed, default
handles.zeroMat(firstNonzero) = max(pic(:));
My input file looks like this:
# FILE:app/src/f1.c
2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 3 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 3 0 0 0 0 0 7 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 24 32 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0
# FILE:src/f2.c
1 0 0 0 0 0 0 0 0 0 0 0 1 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 3 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 0 0 0 0 0 6 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 24 31 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0
...............
I want to read the headers i.e., the lines which start with #, and the vectors present below the headers into lists. I tried the using importdata() as below. But this gives me only the first header and first vector. I need to read it till the end of the file.
filename = 'output.txt';
A = importdata(filename);
disp(A.rowheaders);
disp(A.data);
EDIT: A = importdata(filename,'#'); This solved my problem
I believe the fgetl command should do the trick.
I would put it in a for loop with two fgetl commands, one for the metadata line and one for the numerical data line.
EDIT: Added example
fid = fopen('test.txt');
celHeaders = {};
celData = {};
while(~feof(fid))
celHeaders{end+1} = fgetl(fid);
celData{end+1} = fgetl(fid);
end
fclose(fid);
disp(celHeaders)
disp(celData)
This is assuming that your text file doesn't include additional lines above or below the data.
I have the following input file
-------------------------------------
--- A-TKE spectrum (post-neutron)---
-------------------------------------
2-dim. array: (A = 73 To 162 Step 1) (E = 122 To 198 Step 1)
(The data are written according to the loop structure specified above.
The last loop is the inner-most one. Line breaks are not related to the data structure!
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 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 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 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 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 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 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 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 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 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 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 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 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 1 0 1 1 0 0 0 1 0 0 2 0 0 1 0 0 0 0
</A_TKE>
</Results>
This file is part of a bigger file, which contains more number block similar to the above one.
What I want to do is print in a seperate file the numbers that are within the file. As a first try, I thought "Let's print what is between two patters". The START pattern will be --- A-TKE spectrum (post-neutron)--- and the END pattern will </A_TKE>
I used
awk '/--- A-TKE spectrum (post-neutron)---/{flag=1;next}/</A_TKE>/{flag=0}flag{print}' input
To my surprise nothing is printed on the screen. Any idea on why is this happening?
I would also like to now if I can substitue next with a number that will correspond to the line that I want to print after the given pattern.
In perl (because whilst you start with awk, this is tagged as perl too):
perl -ne 'print if m/post-neutron/...m/A_TKE/' input
Which will do what your thingy was trying to.
It works by using the range operator - which is a perl construct which tests if the current file is between two markers.
And the fact that perl can run inline in the same way as awk or sed. -ne says "wrap this code in a while loop" that iterates a line at a time. (e.g. as sed/awk/grep do by default).
Otherwise ... it depends a little on what else is in your file. Because from the sample you gave - the only lines starting with a number are the ones you want. So:
perl -ne 'print if m/^\s*\d+\s*\d+/'
(Or combine the two tests:
perl -ne 'print if m/post-neutron/...m/A_TKE/ and m/^\s*\d+\s*\d+/'
Which will output anything between the two text markers that has number-space-number at the start of line.
Your awk script is almost there. I'd build it like this:
awk '
/<\/A_TKE>/ {exit}
flag && NF==50 {print}
/--- A-TKE spectrum .post-neutron.---/ {flag=1}
' input
You could optimize it of course, but just looking at the logic of it, the idea is that you need do no further processing once you hit your end marker .. and you only want to start printing AFTER your start marker.
Note that I've replaced the brackets in your start marker with dots, since brackets are interpreted as PART OF THE REGULAR EXPRESSION. I believe you may be considering these simple strings to search for, but they are regex.
Some working ideas
Awk can print between two matching lines with this simple command:
awk '/PATTERN1/,/PATTERN2/' file.txt
If you want more control over what happens when the lines are found, you can use the flag approach.
awk '/A-TKE/{f=1;for (i=0;i<7;i++){getline}}/A_TKE/{f=0}f' file.txt
Adjust i<7 to specify how many lines you would like to skip.
Also, in this particular case, you can exploit the fact that you have 50 fields of numbers, which is unlikely to occur anywhere else in the file. Thus, you could also get the numbers of interest out with:
awk 'NF>40' file.txt
NF is a special awk variable that stores the number of fields (columns) in the current record (line).
Why your approach does not work
Your approach does not work, because everything that is between // in awk is treated as a regular expression. In your case you have several - and () in the string, which are not escaped, and thus assume their special meaning and therefore the expression you typed never matches, thus the flag is never set to 1 and awk does not print a line.
Try this simple sed command
sed -n '/A-TKE/,/<\/A_TKE>/p' FileName
If you want to extract rows with just numbers
awk 'NF && !/[a-zA-Z\-]+/' input.file
will generate
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 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 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 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 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 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 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 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 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 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 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 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 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 1 0 1 1 0 0 0 1 0 0 2 0 0 1 0 0 0 0
with patterns and smart counters you can write it as
awk '/\/A_TKE/{exit}
f&&s&&s--{next}
f&&NF{print}
/A-TKE spectrum/{f=1;s=7}' input.file
fine tune patterns as you like.
So I tried the matrix exponential function using MATLAB's Coder toolkit, and I got it to build. I went on to test to see if the results were reliable and more efficient. While the code was faster, the answer it produced was very slight.
I ran the original function and got an answer of:
p =
1 0 0 0 0 0 0 0 0 0 0 0
-0.05 1 -1.25e-07 5e-06 0 0 0 0 0 0 0 0
0 0 1 0 0 0 0 0 0 0 0 0
1.25e-07 -5e-06 -0.05 1 0 0 0 0 0 0 0 0
0 0 0 0 1 0 0 0 0 0 0 0
0 0 0 0 -0.05 1 0 0 0 0 0 0
-2.0833e-05 0.00125 -5.2083e-11 4.1667e-09 0 0 1 0.05 0 1.25e-07 0 0
-0.00125 0.05 -4.1667e-09 2.5e-07 0 0 0 1 0 5e-06 0 0
5.2083e-11 -4.1667e-09 -2.0833e-05 0.00125 0 0 0 -1.25e-07 1 0.05 0 0
4.1667e-09 -2.5e-07 -0.00125 0.05 0 0 0 -5e-06 0 1 0 0
0 0 0 0 -2.0833e-05 0.00125 0 0 0 0 1 0.05
0 0 0 0 -0.00125 0.05 0 0 0 0 0 1
And I then ran the mexed version of the function with the same input:
p2 =
1 0 0 0 0 0 0 0 0 0 0 0
-0.05 1 -1.25e-07 5e-06 0 0 0 0 0 0 0 0
0 0 1 0 0 0 0 0 0 0 0 0
1.25e-07 -5e-06 -0.05 1 0 0 0 0 0 0 0 0
0 0 0 0 1 0 0 0 0 0 0 0
0 0 0 0 -0.05 1 0 0 0 0 0 0
-2.0833e-05 0.00125 -5.2083e-11 4.1667e-09 0 0 1 0.05 0 1.25e-07 0 0
-0.00125 0.05 -4.1667e-09 2.5e-07 0 0 0 1 0 5e-06 0 0
5.2083e-11 -4.1667e-09 -2.0833e-05 0.00125 0 0 0 -1.25e-07 1 0.05 0 0
4.1667e-09 -2.5e-07 -0.00125 0.05 0 0 0 -5e-06 0 1 0 0
0 0 0 0 -2.0833e-05 0.00125 0 0 0 0 1 0.05
0 0 0 0 -0.00125 0.05 0 0 0 0 0 1
At first glance, these two matrices are equal, but they are actually VERY SLIGHTLY off:
p-p2
ans =
0 0 0 0 0 0 0 0 0 0 0 0
-6.9389e-18 0 0 8.4703e-22 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 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 0 0
-3.3881e-21 -2.1684e-19 -3.2312e-26 8.2718e-25 0 0 0 0 0 5.294e-23 0 0
-2.1684e-19 0 -8.2718e-25 5.294e-23 0 0 0 0 0 8.4703e-22 0 0
6.4623e-27 0 -3.3881e-21 2.1684e-19 0 0 0 0 0 0 0 0
0 0 0 6.9389e-18 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 0 0 0 0 0
Most of the result is equivalent to the original function, but some of it is not. Also, the difference between the two is so small that I cant possibly believe it would be a mathematical error, rather than perhaps a precision error. And the reason I am so concerned with this, is because this does cause issues with the overlaying reason I am using the function.
Is there a reason why the mex function is off by so little, and is there a way to fix this?
The differences you observe are so little that you can consider the results actually are the same.
The way they are computed are different and that is why you do not get exactly the same result. Yet the difference is roughly machine epsilon and is just due to the fact that computers do not work with an infinite precision but with some discrete representation of the numbers.