search

bmihaljevic / bnclassify

Learning Discrete Bayesian Network Classifiers from Data
18 stars 10 forks source link

Error: "are_factors(dataset) is not TRUE" #24

Open blueskypie opened 8 years ago

blueskypie commented 8 years ago

I got that error when apply nb on features of continuous variables. Does nb only work on discrete variables?

Thanks!

bmihaljevic commented 8 years ago

Hi and thanks for asking!

Yes, currently it does support only discrete variables. I have most of the code for including discretization, as I had this working in a previous version, but it needs a bit of polishing. Also, the discretization method itself is rather basic and fails often. This is something I want to add but I currently do not have the time. If you are willing to play with this and include this discretization code into the package I'll pass it to you and give you a couple of guidelines on how I think it should be integrated.

blueskypie commented 8 years ago

Thanks for the quick response! I'm not sure I would have the knowledge to help you. May I take a look at first? or publish it in Github for others to chip in.

Thanks!

bmihaljevic commented 8 years ago

Sorry for not replying earlier. Will publish this code on Github during next week.

psforestry commented 6 years ago

Can I use the code as it exists now to classify a satellite image with numeric integer attributes?

bmihaljevic commented 6 years ago

Currently the only way it to convert them to factors, which is probably not very useful.