Zebin YANG

It seems this can be simply fixed by adding "callbacks=callbacks," to line 2267 of pygam.py.

Hi, We provide a new version of aletheia under the name aletheia-dnn (v1.2). Each plot function will return a plt.figure object, and you can save it by, e.g., fig =...

Hi, Thanks for letting us know this issue. It seems to be a problem related to OS. Our last trial version is compiled for Linux (CentOS) and a new one...

What is your os platform?

I compiled a mac version and uploaded it to pipy. You may have another try.

Hi, A new version is compiled for py36, py37, and py38. try “pip install aletheia-dnn”

Hi, It seems to be related to the mismatch of your X_train and weights, i.e., the number of input features (X_train.shape[1]) does not match the number of rows of the...

Another possible reason is that if your response varable y has more than one column, then you will also get an error. Currently, the package only supports binary classification.

That is because your weight list and bias list have different lengths.

Hi @virtualitics-joseph, this is an error due to lightgbm on MacOS. You may reinstall lightgbm using the solution in the following link. https://stackoverflow.com/questions/50889162/lightgbm-image-not-found-after-mac-install