Column with real numbers identified as integer type
iar73
New Altair Community Member
I have version 5.0.003. A data column in the csv file contains the following values: 0.0001 and 1.0001 RapidMiner recognizes the attribute as type INTEGER instead of REAL. However, values are handled correctly (they are not converted to integer).
Looking at the source I think the problem is here, with the rounding:
CSVDataReader(AbstractDataReader).init() line: 266
if (canParseInteger) {
if (!Tools.isEqual(Math.round(number.doubleValue()), number.intValue())) {
canParseInteger = false;
}
}
Best regards,
András
Looking at the source I think the problem is here, with the rounding:
CSVDataReader(AbstractDataReader).init() line: 266
if (canParseInteger) {
if (!Tools.isEqual(Math.round(number.doubleValue()), number.intValue())) {
canParseInteger = false;
}
}
Best regards,
András
0
Answers
-
Hi,
input operators have been overhauled for 5.1. The line of code you mention is no longer there.
Cheers,
Simon0