File tree Expand file tree Collapse file tree 1 file changed +1
-4
lines changed Expand file tree Collapse file tree 1 file changed +1
-4
lines changed Original file line number Diff line number Diff line change @@ -147,10 +147,7 @@ in the kernel and by the regularization parameter alpha of KRR.
147
147
:align: center
148
148
149
149
The figure shows that both methods learn reasonable models of the target
150
- function. GPR correctly identifies the periodicity of the function to be
151
- roughly :math: `2 *\pi ` (6.28), while KRR chooses the doubled periodicity
152
- :math: `4 *\pi ` . Besides
153
- that, GPR provides reasonable confidence bounds on the prediction which are not
150
+ function. GPR provides reasonable confidence bounds on the prediction which are not
154
151
available for KRR. A major difference between the two methods is the time
155
152
required for fitting and predicting: while fitting KRR is fast in principle,
156
153
the grid-search for hyperparameter optimization scales exponentially with the
You can’t perform that action at this time.
0 commit comments