I have a problem

Question

SULE SAHIN 2017-11-6

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/365389-i-have-a-problem

评论： Walter Roberson 2018-4-27

Write a function called integerize that takes as its input a matrix A of integers of type double, and returns the name of the “smallest” signed integer class to which A can be converted without loss of information. If no such class exists, the text 'NONE' is returned. For example, if the smallest element of A is -100 and the largest is +100, then the function would return 'int8'. As another example, if there is an element of A equal to -1e20, then the function would return 'NONE'.

My code is;

if maxA<=((2^7)-1) && minA>=-(2^7);
    x='int8';
  elseif maxA<=((2^15)-1) && maxA>((2^7)-1) && minA>=-(2^15) && minA< -(2^7);
      x='int16';
  elseif maxA<=((2^31)-1) && maxA>((2^15)-1) && minA >=-(2^31) && minA <-(2^15);
      x='int32';
  elseif maxA<=((2^63)-1) && maxA>((2^31)-1) && minA >=-(2^63) && minA< -(2^31);
    x='int64';
elseif maxA<= ((2^8)-1) && minA >= 0;
    x = 'uint8';
elseif maxA<= ((2^16)-1) && maxA >((2^8)-1);
    x = 'uint16';
elseif maxA<= ((2^32)-1) maxA > ((2^16)-1) ;
    x = 'uint32';
elseif maxA<= ((2^64)-1) && maxA > ((2^32)-1) ;
    x = 'uint64';
else
      x=NONE;
  end
end

But it is not correct. Where is my error?

1 个评论
显示 -1更早的评论隐藏 -1更早的评论

Walter Roberson 2018-4-27

Please do not close questions that have an answer.

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

Jan 2017-11-6

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/365389-i-have-a-problem#answer_289690

编辑：Jan 2017-11-6

在 MATLAB Online 中打开

It is asked for "signed integer". Then uint32 is not an option, because it is unsigned.

x = 'NONE'; % requires quotes

The operator is missing in:

elseif maxA<= ((2^32)-1)   maxA > ((2^16)-1) ;
                        ^^

In addition this does not restrict the minA part.

You do not have to exclude e.g. the int8 range, if it was excluded already:

if maxA<=((2^7)-1) && minA>=-(2^7);
   x = 'int8';
elseif maxA<=((2^15)-1) && maxA>((2^7)-1) && minA>=-(2^15) && minA< -(2^7);
    x='int16';

Easier:

if maxA <= 2^7 - 1 && minA >= -2^7
   x = 'int8';
elseif maxA <= 2^15 - 1 && minA >= -2^15
   x = 'int16';

I'd prefer a loop:

TypeList = {'int8', 'uint8', 'int16', 'uint16', ...
            'int32', 'uint32', 'int64', 'uint64'};
x = 'NONE';
for k = 1:numel(TypeList)
  if isequal(double(A), double(cast(A, TypeList{k})))
    x = TypeList{k};
    break;
  end
end

Or:

minA = min(A(:));
maxA = max(A(:));
x    = 'NONE';
for aType = {'int8', 'int16', 'int32', 'int64'}
  aType = TypeList{k}
  if minA >= intmin(aType) && maxA <= intmax(aType)
     x = aType;
     break;
  end
end

Or

minA = min(A(:));
maxA = max(A(:));
x = 'NONE';
for k = [8,16,32,64]
  if minA >= -2^(k-1) && maxA <= 2^(k-1)
     x = sprintf('int%d', k);
     break;
  end
end

3 个评论
显示 1更早的评论隐藏 1更早的评论

SULE SAHIN 2017-11-6

在 MATLAB Online 中打开

function x = integerize(A) maxA = max(A(:)); minA = min(A(:));

if maxA<=((2^7)-1) && minA>=-(2^7);
    x='int8';
  elseif maxA<=((2^15)-1) && minA>=-(2^15); 
      x='int16';
  elseif maxA<=((2^31)-1) && minA >=-(2^31) ;
      x='int32';
  elseif maxA<=((2^63)-1) && minA >=-(2^63) ;
    x='int64';
elseif maxA<= ((2^8)-1) && minA >= 0;
    x = 'uint8';
elseif maxA<= ((2^16)-1) && minA >= 0;
    x = 'uint16';
elseif maxA<= ((2^32)-1) &&  minA >= 0;
    x = 'uint32';
elseif maxA<= ((2^64)-1) &&  minA >= 0;
    x = 'uint64';
else
      x='NONE';
  end
end