Would be great if this could be made compatible with Python 3.
I tried to do this, but I am now stuck with the pickled models where the loader cannot unpickle with the default encoding "ascii" nor with "utf8". Unpickling with "latin1" works, but I am not sure why and the Sklearn model is version 0.18.1 which may not work with 0.19.1 or later, apparently.
Is there any simple way to make this work in Python 3?
Would be great if this could be made compatible with Python 3. I tried to do this, but I am now stuck with the pickled models where the loader cannot unpickle with the default encoding "ascii" nor with "utf8". Unpickling with "latin1" works, but I am not sure why and the Sklearn model is version 0.18.1 which may not work with 0.19.1 or later, apparently.
Is there any simple way to make this work in Python 3?