search

wjmaddox / online_gp

Code repo for "Kernel Interpolation for Scalable Online Gaussian Processes"
Apache License 2.0
54 stars 7 forks source link

Adaption to high-dimentional input such as 3d input #4

Open trannguyenle95 opened 6 months ago

trannguyenle95 commented 6 months ago

Hi,

I have been trying to adapt our code to 3d input but so far failed to do so. Is it possible to give hints regarding this matter? Thanks a lot!

cagoc commented 2 months ago

Hi, Sorry to bother you. I have been trying to reproduce active_learning.py with wiski model but I failed. it is said that

Traceback (most recent call last): File "C:\code\online_vargp-main\experiments\active_learning\qnIPV_experiment.py", line 308, in <module> main(args) File "C:\code\online_vargp-main\experiments\active_learning\qnIPV_experiment.py", line 146, in main mll = mll_type(model.likelihood, model, **mll_kwargs) ^^^^^^^^^^^^^^^^ AttributeError: 'NoneType' object has no attribute 'likelihood' I have tried to use ExactMarginalLogLikelihood and BatchedWoodburyMarginalLogLikelihood but none of them succeeded. Have you reproduce active_learning.py with wiski model successfully? Thanks a lot!