Breaking change: switched position of y_true and y_predicted in p_fun to match the keras metrics methods.
* Gb/gan (20) grantbuster
Bug Fixes
* add shuffle to model interfaces (19) MRossol
0.0.7
What’s Changed
phygnn wasnt executing dropout layers correctly. The Dropout layer object needed a boolean flag to know if it was training or not. Fixed with additional tests.
* Gb/dropout bug (18) grantbuster
0.0.6
What’s Changed
* phygnn save() to now save optimizer config dict instead of optimizer … (17) grantbuster
0.0.5
What’s Changed
* Ohe norm fix (15) MRossol * Allow changing loss weights on phygnn model interface (13) mikebannis * Allow RNG seeding before model initiation (12) mikebannis
Features
* add one_hot_categories to model handlers (14) MRossol
Bug Fixes
* add one_hot_categories to model handlers (14) MRossol
0.0.4
What’s Changed
Added phygnn model as input to p_fun. This will enable custom loss functions that can evaluate predictions of the model at the current moment in training.
Added dummy pfun for user example.
* Added phygnn model instance as input arg to p_fun. Added dummy exaple (11) grantbuster