DOC Added example link to Gaussian process regressor #26980
Conversation
sklearn/kernel_ridge.py
Outdated
| @@ -129,6 +129,9 @@ class KernelRidge(MultiOutputMixin, RegressorMixin, BaseEstimator): | |||
| >>> krr = KernelRidge(alpha=1.0) | |||
| >>> krr.fit(X, y) | |||
| KernelRidge(alpha=1.0) | |||
|
|
|||
| See :ref:`sphx_glr_auto_examples_gaussian_process_plot_compare_gpr_krr.py` | |||
| for an example to compare kernel ridge and Gaussian process regression | |||
There was a problem hiding this comment.
I looked at the documentation of the KernelRidge and we have a few examples (three of them). I think that the title currently is self-explicit and we don't need to add a new cross-reference here.
There was a problem hiding this comment.
I removed the example from KernelRidge, can you please review? Thanks!
sklearn/gaussian_process/_gpr.py
Outdated
|
|
||
| See :ref:`sphx_glr_auto_examples_gaussian_process_plot_compare_gpr_krr.py` | ||
| for an example to compare kernel ridge and Gaussian process regression |
There was a problem hiding this comment.
I would move this reference to the general summary at the top of the file just before mentioning the user-guide.
I think that this example is actually didactic to understand the difference between a frequentist approach (KernelRidge) and the more Bayesian approach (GaussianProcessRegressor).
I think that we should rephrase as:
To learn the difference between a point-estimate approach vs. a more
Bayesian modelling approach, refer to the example entitled
:ref:`sphx_glr_auto_examples_gaussian_process_plot_compare_gpr_krr.py`.There was a problem hiding this comment.
I made the suggested changes in GaussianProcessRegressor, can you please review? Thanks!
mecopur
force-pushed
the
plot_compare_gpr_krr
branch
from
2173849
to
62dee06
Compare
mecopur
changed the title
Reference Issues/PRs:
26927
What does this implement/fix? Explain your changes.
Added example plot_compare_gpr_krr.py link to kernel_ridge.py and _gpr.py