I'm trying to build a matrix out of linear equations but for some reason I keep getting a parse error in my matrix when previously I did not.
CoM=[K1*(abs(Q1).^r) K2*(abs(Q2).^r) -(K3*(abs(Q3).^r)) -(K4*(abs(Q4).^r)); K3*(abs(Q3).^r) -(K5*(abs(Q5).^r)) -(K6*(abs(Q6).^r) -(K7*(abs(Q7).^r) ; 1 -1 0 0 ; 1 1 -1 0 ; 1 -1 0 0 ; 1 -1 0 0 ; 1 1 -1 0];
^
error: parse error near line 1 of file '____'\WNA2loop.m
syntax error
CoM=[K1*(abs(Q1).^r) K2*(abs(Q2).^r) -(K3*(abs(Q3).^r)) -(K4*(abs(Q4).^r)); K3*(abs(Q3).^r) -(K5*(abs(Q5).^r)) -(K6*(abs(Q6).^r) -(K7*(abs(Q7).^r) ; 1 -1 0 0 ; 1 1 -1 0 ; 1 -1 0 0 ; 1 -1 0 0 ; 1 1 -1 0];
^
Right where I put the caret but when I take that small part of the matrix and run the command it
[1 -1 0 0 ; 1 1 -1 0 ; 1 -1 0 0 ; 1 -1 0 0 ; 1 1 -1 0]
ans = 1 -1 0 0
1 1 -1 0
1 -1 0 0
1 -1 0 0
1 1 -1 0
The K1 to K7 and Q1 to Q7 as well as the are just variables that get input by the user through the input function it worked before but now it just wont budge, could someone please provide assistance on this?
Related
I have a program that prints out the following:
bash-3.2$ ./drawgrid
0
1 1 0
1 1 0
0 0 0
1
0 1 1
0 1 1
0 0 0
2
0 0 0
1 1 0
1 1 0
3
0 0 0
0 1 1
0 1 1
Is it possible to pipe the output of this command such that I get all the 3x3 matrices (together with their number) displayed on a table, for example a 2x2 like this?
0 1
1 1 0 0 1 1
1 1 0 0 1 1
0 0 0 0 0 0
2 3
0 0 0 0 0 0
1 1 0 0 1 1
1 1 0 0 1 1
I tried searching, and came across the column command, but I did not figure it out.
Thank you
You can use pr -2T to get the following output, which is close to what you expected:
0 2
1 1 0 0 0 0
1 1 0 1 1 0
0 0 0 1 1 0
1 3
0 1 1 0 0 0
0 1 1 0 1 1
0 0 0 0 1 1
You could use an awk script:
NF == 1 {
if ($NF % 2 == 0) {
delete line
line[1]=$1
f=1
} else {
print line[1]"\t"$1
f=0
}
n=1
}
NF > 1 {
n++
if (f)
line[n]=$0
else
print line[n]"\t"$0
}
And pipe to it like so:
$ ./drawgrid | awk -f 2x2.awk
0 1
1 1 0 0 1 1
1 1 0 0 1 1
0 0 0 0 0 0
2 3
0 0 0 0 0 0
1 1 0 0 1 1
1 1 0 0 1 1
You can get exactly what you expect with a short bash script and a little array index thought:
#!/bin/bash
declare -a idx
declare -a acont
declare -i cnt=0
declare -i offset=0
while IFS=$'\n'; read -r line ; do
[ ${#line} -eq 1 ] && { idx+=( $line ); ((cnt++)); }
[ ${#line} -gt 1 ] && { acont+=( $line );((cnt++)); }
done
for ((i = 0; i < ${#idx[#]}; i+=2)); do
printf "%4s%8s\n" ${idx[i]} ${idx[i+1]}
for ((j = offset; j < offset + 3; j++)); do
printf " %8s%8s\n" ${acont[j]} ${acont[j+3]}
done
offset=$((j + 3))
done
exit 0
Output
$ bash array_cols.sh <dat/cols.txt
0 1
1 1 0 0 1 1
1 1 0 0 1 1
0 0 0 0 0 0
2 3
0 0 0 0 0 0
1 1 0 0 1 1
1 1 0 0 1 1
I have a file with lots of pieces of information that I want to split on the first column.
Example (example.gen):
1 rs3094315 752566 A G 0 1 0 1 0 0 1 0 0 0 1 0 0 1
1 rs2094315 752999 A G 0 1 0 1 0 0 1 0 0 0 1 0 0 1
2 rs3044315 759996 A G 0 1 0 1 0 0 1 0 0 0 1 0 0 1
2 rs3054375 799966 A G 0 1 0 1 0 0 1 0 0 0 1 0 0 1
2 rs3094375 999566 A G 0 1 0 1 0 0 1 0 0 0 1 0 0 1
3 rs3078315 799866 A G 0 1 0 1 0 0 1 0 0 0 1 0 0 1
3 rs4054315 759986 A G 0 1 0 1 0 0 1 0 0 0 1 0 0 1
4 rs4900215 752998 A G 0 1 0 1 0 0 1 0 0 0 1 0 0 1
4 rs5094315 759886 A G 0 1 0 1 0 0 1 0 0 0 1 0 0 1
4 rs6094315 798866 A G 0 1 0 1 0 0 1 0 0 0 1 0 0 1
Desired output:
Chr1.gen
1 rs3094315 752566 A G 0 1 0 1 0 0 1 0 0 0 1 0 0 1
1 rs2094315 752999 A G 0 1 0 1 0 0 1 0 0 0 1 0 0 1
Chr2.gen
2 rs3044315 759996 A G 0 1 0 1 0 0 1 0 0 0 1 0 0 1
2 rs3054375 799966 A G 0 1 0 1 0 0 1 0 0 0 1 0 0 1
2 rs3094375 999566 A G 0 1 0 1 0 0 1 0 0 0 1 0 0 1
Chr3.gen
3 rs3078315 799866 A G 0 1 0 1 0 0 1 0 0 0 1 0 0 1
3 rs4054315 759986 A G 0 1 0 1 0 0 1 0 0 0 1 0 0 1
Chr4.gen
4 rs4900215 752998 A G 0 1 0 1 0 0 1 0 0 0 1 0 0 1
4 rs5094315 759886 A G 0 1 0 1 0 0 1 0 0 0 1 0 0 1
4 rs6094315 798866 A G 0 1 0 1 0 0 1 0 0 0 1 0 0 1
I've tried to do this with the following shell scripts, but it doesn't work - I can't work out how to get awk to recognise a variable defined outside the awk script itself.
First script attempt (no awk loop):
for i in {1..23}
do
awk '{$1 = $i}' example.gen > Chr$i.gen
done
Second script attempt (with awk loop):
for i in {1..23}
do
awk '{for (i = 1; i <= 23; i++) $1 = $i}' example.gen > Chr$i.gen
done
I'm sure its probably quite basic, but I just can't work it out...
Thank you!
With awk:
awk '{print > "Chr"$1".gen"}' file
It just prints and redirects it to a file. And how is this file defined? With "Chr" + first_column + ".gen".
With your sample input it creates 4 files. For example the 4th is:
$ cat Chr4.gen
4 rs4900215 752998 A G 0 1 0 1 0 0 1 0 0 0 1 0 0 1
4 rs5094315 759886 A G 0 1 0 1 0 0 1 0 0 0 1 0 0 1
4 rs6094315 798866 A G 0 1 0 1 0 0 1 0 0 0 1 0 0 1
First, use #fedorqui's answer, as that is best. But to understand the mistake you made with your first attempt (which was close), read on.
Your first attempt failed because you put the test inside the action (in the braces), not preceding it. The minimal fix:
awk "\$1 == $i" example.gen > Chr$i.gen
This uses double quotes to allow the value of i to be seen by the awk script, but that requires you to then escape the dollar sign for $1 so that you don't substitute the value of the shell's first positional argument. Cleaner but longer:
awk -v i=$i '$1 == i' example.gen > Chr$i.gen
This adds creates a variable i inside the awk script with the same value as the shell's i variable.
I am looking for a C++ library for Discrete Wavelet Transform (DWT) which can also return
the NxN DWT matrix of the transform.
There was a similar question opened here
Looking for a good C/C++ wavelet library for signal processing
but I am looking for something more specific as you can see.
It would be more helpful if the library is under some non-GNU license that lets me use it in proprietary software (LGPL, MPL, BSD etc.)
Thanks in advance
The reason why this matrix is never computed is that it is very inefficient to compute the DWT using it. The FWT approach is much faster.
For a signal of length 16 and a 3-level haar transform, I found that this matrix in matlab
>> h=[1 1];
>> g=[1 -1];
>> m1=[[ones(1,8) zeros(1,8); ...
zeros(1,8) ones(1,8); ...
1 1 1 1 -1 -1 -1 -1 zeros(1,8); ...
zeros(1,8) 1 1 1 1 -1 -1 -1 -1]/sqrt(8); ...
[1 1 -1 -1 zeros(1,12); ...
zeros(1,4) 1 1 -1 -1 zeros(1,8); ...
zeros(1,8) 1 1 -1 -1 zeros(1,4); ...
zeros(1,12) 1 1 -1 -1]/sqrt(4); ...
[g zeros(1,14); ...
zeros(1,2) g zeros(1,12); ...
zeros(1,4) g zeros(1,10); ...
zeros(1,6) g zeros(1,8); ...
zeros(1,8) g zeros(1,6); ...
zeros(1,10) g zeros(1,4); ...
zeros(1,12) g zeros(1,2); ...
zeros(1,14) g]/sqrt(2)]
m1 =
A A A A A A A A 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 A A A A A A A A
A A A A -A -A -A -A 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 A A A A -A -A -A -A
B B -B -B 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 B B -B -B 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 B B -B -B 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 B B -B -B
C -C 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 C -C 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 C -C 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 C -C 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 C -C 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 C -C 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 C -C 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 C -C
where A=1/sqrt(8), B=1/sqrt(4) and C=1/sqrt(2).
corresponds to the FWT. That shows you how you build your matrix from the filters. You start with the bottom half of the matrix --a matrix of zeroes, putting filter g 2 steps further every row. then make the filter twice as wide and repeat, only now shift 4 steps at a time. repeat this until you are at the highest level of decomposition, the finally put the approximation filter in at the same width (here, 8).
just as a check
>> signal=1:16; % ramp
>> [h g]=daubcqf(2); % Haar coefficients from the Rice wavelet toolbox
>> fwt(h,signal,3) % fwt code by Jeffrey Kantor
>> m1*signal' % should produce the same vector
Hope that helps you writing it in C++. It is not difficult (a bit of bookkeeping) but as said, noone uses it because efficient algorithms do not need it.
How would I go about creating the matrix
[1 2 0 0 0;
-1 1 2 0 0;
0 -1 1 2 0;
0 0 -1 1 2;
0 0 0 -1 1]
using the diag command in MatLab?
Here is one way:
> diag(ones(1,5),0)+diag(ones(1,4),1)*2+diag(ones(1,4),-1)*-1
ans =
1 2 0 0 0
-1 1 2 0 0
0 -1 1 2 0
0 0 -1 1 2
0 0 0 -1 1
>
This just creates three diagonals at 0, +1 and -1, scales them as needed, then adds them.
I have a binarized image like the folowing matrix:
1 1 0 0 1 1 0 0 1 1 0 0 0 0 1 1 1 1 0 0 1 1 0 0 1 1
1 1 0 0 1 1 0 0 1 1 0 0 0 0 1 1 1 1 0 0 1 1 0 0 1 1
1 1 0 0 1 1 0 0 1 1 0 0 0 0 1 1 1 1 0 0 1 1 0 0 1 1
1 1 0 0 1 1 0 0 1 1 0 0 0 0 1 1 1 1 0 0 1 1 0 0 1 1
The problem is that the image stars and end with 101, so how can i turn that into this.
1 0 1 0 1 0 0 1 1 0 1 0 1
1 0 1 0 1 0 0 1 1 0 1 0 1
1 0 1 0 1 0 0 1 1 0 1 0 1
1 0 1 0 1 0 0 1 1 0 1 0 1
I am trying to the decode the image binary code.
it seems like resizing original image with scale (0.5,1) using Nearest Neighbor method.
If you are using Matlab or any other language with similar array processing capabilities (APL, Fortran 90, Mathematica, C++ +Boost, ...) you could turn your input into your desired output with a statement similar to this:
arr(:,1:2:end)
if your array of pixels is called arr of course.
This does not return the median of close pixels, but then nor does the suggested output in the question.