daniel servén

@hongkahjun unfortunately this is a somewhat deep issue. When writing the new `terms` functionality, i diverged from sklearn's requirement that the estimator instance's parameters are not changed after looking at...

yes! i would love to add that for parallelizing the gridsearch and bootstrapping tasks. i've only messed around with it a little bit with Multiprocessing (https://docs.python.org/2/library/multiprocessing.html) and Pathos (https://github.com/uqfoundation/pathos) but...

for some reason i don't think i've tried joblib... in practice i've run into implementations issues (some stemming from OSX) with the various fork types (`fork`, `forkserver`, `spawn`, etc), executing...

@jeweinberg @h4gen thanks for the tips. i am adding joblib for concurrent execution and out-of-core learning. do you all know if it is necessary to add the `partial_fit()` method for...

@RobbieShan not yet, but that would be cool. im not very familiar with the ordered logit. does it require multiple decision functions the way a catgorical distribution would?

hey @AlexEngelhardt ! sorry for the slowwww response. the way to do this for the general case it to use the empirical sampling approach (there's some discussion [here](https://github.com/dswah/pyGAM/issues/154#issuecomment-366618372)) but here's...

@AlexEngelhardt did this work for you?

@alexengelhardt im very glad it worked for you. this is the correct way to compute prediction intervals for non-gaussian distributions. it definitely belongs in the `prediction_intervals` method for non-gaussian models....

it should be easy to close https://github.com/dswah/pyGAM/issues/156 with this method.

@arose13 this is awesome! thank you very much! However, we have recently changed the license to Apache 2.0 to allow proprietary use of the library, which means downstream users would...