AaltoML/BayesNewton
Bayes-Newton—A Gaussian process library in JAX, with a unifying view of approximate Bayesian inference as variants of Newton's method.
PythonApache-2.0
Issues
- 1
GPU training
#19 opened by zhangn2015 - 1
SparseMarkovVariationalGP vs S2CVI
#30 opened by sethaxen - 5
air_quality.py demo failing
#35 opened by Srujan-D - 1
Input-dependent noise model
#24 opened by itskalvik - 4
- 1
Missing data for air_quality demo
#14 opened by danieliong - 3
- 1
Missing data for experiments, (electricity ...)
#31 opened by mat-ej - 1
energy function can output NaN if compiled
#27 opened by ThoreWietzke - 5
Periodic Kernel outputs wrong period
#22 opened by ThoreWietzke - 0
- 0
kernels.py line 406
#17 opened by mtsaari - 2
- 1
Sparse EP energy is incorrect
#2 opened by wil-j-wil - 1
Double Precision Issues
#13 opened by harrisonzhu508 - 2
jitted predict
#7 opened by soldierofhell - 7
Addition of a Squared Exponential Kernel?
#10 opened by mathDR - 1
How to understand the function `cavity_distribution_tied` in file `basemodels.py` ?
#8 opened by Fangwq - 2
error in heteroscedastic.py with MarkovPosteriorLinearisationQuasiNewtonGP method
#6 opened by Fangwq - 2
How to set initial `Pinf` variable in kernel?
#5 opened by Fangwq - 1
issue with SparseVariationalGP method
#4 opened by Fangwq - 2