Filter löschen
Filter löschen

If statement with many logical or.

1 Ansicht (letzte 30 Tage)
sourabh mittal
sourabh mittal am 8 Okt. 2018
Beantwortet: John D'Errico am 8 Okt. 2018
This is my code. I have a1 = 0.4 and b1 =0.2 and 1-a1-b1 = 0.4. Middle condition of if statement is satisfied but still it doesn't work. please help.
% code
for a1 = 0.4 : 0.1 : 1
for b1 = 0.2 : 0.1 : 1
if (a1 == b1 && b1 > 1-a1-b1) || (1-a1-b1 == a1 && a1 > b1) || (1-a1-b1 == b1 && b1 > a1)
continue;
end
file_name = sprintf('file_%.2f_%.2f_%.2f.txt',a1,b1,1-a1-b1);
fileID = fopen(file_name, 'w+');
fclose(fileID);
end
end
  9 Kommentare
Adam
Adam am 8 Okt. 2018
Yes, looking at the loop again, this could well cause a problem.
sourabh mittal
sourabh mittal am 8 Okt. 2018
% code
for a1 = 0.4 : 0.1 : 1
for b1 = 0.2 : 0.1 : 1
if (1-a1-b1 == a1 && a1 > b1)
continue;
end
file_name = sprintf('file_%.2f_%.2f_%.2f.txt',a1,b1,1-a1-b1);
fileID = fopen(file_name, 'w+');
fclose(fileID);
end
end
See this condition of if statement. this is not working as i described above.

Melden Sie sich an, um zu kommentieren.

Akzeptierte Antwort

Dennis
Dennis am 8 Okt. 2018
You are comparing floating point numbers, which will not necessary work:
1-0.4-0.2==0.4
ans =
0
You can use
abs(1-a1-b1-a1)<0.001
For more information please look here.

Weitere Antworten (2)

John D'Errico
John D'Errico am 8 Okt. 2018
Your problem is EXACTLY in that you are trying to compare floating point numbers, for exact equality. This is something you should never do, unless you completely understand floating point numbers, and how to know when you can do it safely. And even then you still need to take extra care.
But tests like this:
b1 > 1-a1-b1
are just a flat out bad idea. NEVER DO THAT. NEVER. Must I say it again? Don't do it. Period. For example, just how does 0.4 compare to 1-0.2-0.4? Surely they are the same number?
sprintf('%0.55f',0.4)
ans =
'0.4000000000000000222044604925031308084726333618164062500'
sprintf('%0.55f',1 - 0.4 - 0.2)
ans =
'0.3999999999999999666933092612453037872910499572753906250'
Not so.
Instead, you do have options.
for a1i = 4 : 10
a1 = a1i/10;
for b1i = 2 : 10
b1 = b1i/10;
if (a1i == b1i && b1i > 10-a1i-b1i) || (10-a1i-b1i == a1i && a1i > b1i) || (10-a1i-b1i == b1i && b1i > a1i)
yada yada yada...
As you should see, I changed all of your numbers to be integers, double precision integers, but still integers. All of the tests (assuming I did not mistype anything in that last line) work on INTEGERS, thus are comparisons between integers.
Compute a1 and b1 separately. Use them in floating point form, but NEVER use them for comparisons. NEVER. Well, again, unless you seriously know what you are doing, and you know why you should never use tests of that form.

ANKUR KUMAR
ANKUR KUMAR am 8 Okt. 2018
Write statement inside of the loop
for a1 = 0.4 : 0.1 : 1
for b1 = 0.2 : 0.1 : 1
if (a1 == b1 && b1 > 1-a1-b1) || (1-a1-b1 == a1 && a1 > b1) || (1-a1-b1 == b1 && b1 > a1)
file_name = sprintf('file_%.2f_%.2f_%.2f.txt',a1,b1,1-a1-b1);
fileID = fopen(file_name, 'w+');
fclose(fileID);
end
end
end

Kategorien

Mehr zu Loops and Conditional Statements finden Sie in Help Center und File Exchange

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by