I am writing a script to do some classification using weka and I get an error when I try to run the classifier. I have converted the file from CSV using weka.core.converters.CSVLoader.
I have then removed some of the attributes from it using weka.filters.unsupervised.attribute.Remove -R 1,7,8
Now when I try to run the classifier on it using the command
java -classpath /home/jim/Downloads/weka-3-6-3/weka.jar weka.classifiers.trees.J48 -p 0 -split-percentage 66 -preserve-order -t arff_input/Config38-R.arff
I get an error saying
java.io.IOException: Unable to determine structure as arff (Reason: java.io.IOException: premature end of file, read Token[EOF], line 1).
EDIT - I presume this is an error in my first line so here it is.
@relation 'Config38-weka.filters.unsupervised.attribute.Remove-R1,开发者_如何学编程7,8'
Has anyone else encountered a similar problem? Or have I made a mistake some where.
Thanks
I changed the first line to
@relation Config38
and now it works. No idea why this works and the above didn't I'm afriad
If you delete the first line in .arff file it will also work.
精彩评论