Data Science Stack Exchange is a question and answer site for Data science professionals, Machine Learning specialists, and those interested in learning more about the field. It only takes a minute to sign up.
Sign up to join this community
I'm trying to get used to text mining using orange. I set up a very easy workflow loading a corpus (tab separated text file containing 10 entries), using a corpus viewer that shows my data correctly and a topic discovery item. But I always receive an error Message at 95% of topic discovery called:
Message: "Error calling '%s' of '%s'"
Arguments: ('set_data', 'Topic Discovery')
What do you think am I doing wrong here? Thank you very much for your help in advance.
Stefan
One thing I've noticed about Orange is that anytime you get an error calling set_data, there will be some sort of ValueError or FormatError that it is a child of.
Take a look at the rest of the errors and look for some type of value or formatting error, or post your full error log in a pastebin so I can take a look.
Which version are you running? Does the same happen on the 0.2.2 version released a few days ago (note that for 0.2.2 you need also the newest version of Orange; i.e 3.3.9)?
It looks like you're not doing anything wrong but rather that this is a bug. If you could paste the whole error log, perhaps also a schema and a sample of the data if possible, along with the version of Orange and Orange3-Text that you are running I promise to take a look and fix the bug.
