You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
In Fast Predictions it is noted that parts of the predictive mean and variance do not depend on the test data, hence they can be pre-computed.
Using the posterior = model.posterior() I can use the cached method of predicting using posterior.predict_f. Also in the posterior.cache attribute there are 3 elements 2 lower triangular matrices and 1 vector of length m where m is my number of inducing points. I would think the vector is the alpha referred to in the predictive mean, whereas the other two matrices seem to be the cholesky decomposition of some matrix or some lower triangular matrices. Makes me think that the cached quantities need to be combined in a way to obtain the predictive mean and covariance. Any insight on what these are in the context of SGPR?
reacted with thumbs up emoji reacted with thumbs down emoji reacted with laugh emoji reacted with hooray emoji reacted with confused emoji reacted with heart emoji reacted with rocket emoji reacted with eyes emoji
-
In Fast Predictions it is noted that parts of the predictive mean and variance do not depend on the test data, hence they can be pre-computed.
Using the
posterior = model.posterior()
I can use the cached method of predicting usingposterior.predict_f
. Also in theposterior.cache
attribute there are 3 elements 2 lower triangular matrices and 1 vector of lengthm
wherem
is my number of inducing points. I would think the vector is the alpha referred to in the predictive mean, whereas the other two matrices seem to be the cholesky decomposition of some matrix or some lower triangular matrices. Makes me think that the cached quantities need to be combined in a way to obtain the predictive mean and covariance. Any insight on what these are in the context of SGPR?Beta Was this translation helpful? Give feedback.
All reactions