Hi,
in the first picture below you see the problematic bubble chart:
Too many items in the middle of the screen.
Log scaling is not possible because of negative values on both axis.
To get a better distribution of the bubbles, I made a log transformation of the original values by simply taking the logarithm of each value:
for( my $i = 0; $i < @datasets;$i++)
{
for(my $j = 0; $j < scalar(@{$datasets[$i]});$j++)
{
my $v = $datasets[$i]->[$j];
if( $v != 0 )
{
my $s = $v < 0 ? -1 : 1 ;
$v = log(abs($v))*$s;
$datasets[$i]->[$j] = $v
}
}
}
The second picture shows the values after that transformation.
Now the question is: What possibilities do I have in CD to get the right (original) numbers on the axes - best would be something like:
-100,000,000; -10,000; -1,000; -100; -10, -1; 0; 1; 10; ....
Regards
Sinan
|