What stack can handle some data preprocessing and output in ARFF format?
I used to do my preprocessing of data in Python, loading CSV or other things and constructing a specific program for each problem I had to address where an output in a ARFF format was needed.
Of course Pentaho's Data Integration Tool (formerly know as Keetle) should work for that kind of tasks, but it's not. There's no ARFF export build-in available, I've been lost by the trick that should be run to get this working.
The Weka integrated Knowledge Flow crashes on my CSV (fields names are too long), so I might have to go back to Python.
Is there anything that should do the trick, that I don't know ?
I used to do my preprocessing of data in Python, loading CSV or other things and constructing a specific program for each problem I had to address where an output in a ARFF format was needed.
Of course Pentaho's Data Integration Tool (formerly know as Keetle) should work for that kind of tasks, but it's not. There's no ARFF export build-in available, I've been lost by the trick that should be run to get this working.
The Weka integrated Knowledge Flow crashes on my CSV (fields names are too long), so I might have to go back to Python.
Is there anything that should do the trick, that I don't know ?
No comments:
Post a Comment