search

SheffieldML / GPy

Gaussian processes framework in python
BSD 3-Clause "New" or "Revised" License
2.04k stars 562 forks source link

Assertion error #892

Open rfagyei opened 3 years ago

rfagyei commented 3 years ago

Hello

I am trying to run the GPy classification with a dataset where I am sampling training and testing datasets (8-fold cross validation). I encountered an assertion error and to which some assistance will be very appreciated.

The error reads below as:

assert x > 0.0, "Arg of erfRationalHelper should be >0.0; was {}".format(x) AssertionError: Arg of erfRationalHelper should be >0.0; was nan

Thank you. Sincerely,

Ronald Agyei

ekalosak commented 3 years ago

Hey rfagyei, can you post the full traceback? It looks like, I'm guessing from this minimal impression, you're getting some training or testing data that has NaNs, and the erf (error function, i.e. the integral of the normal distribution from -infty to x) can't handle NaN input.