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
Copy file name to clipboardExpand all lines: source/content/models/CommitteeRegressor.rst
+1-1Lines changed: 1 addition & 1 deletion
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -16,7 +16,7 @@ Measuring disagreement
16
16
17
17
With an ensemble of regressors like in the CommitteeRegressor model, a measure of disagreement can be the standard deviation of predictions, which provides a simple way to query for labels. This is not the case in general: for ordinary regressors, it is difficult to come up with viable query strategies because they don't always provide a way to measure uncertainty. (One notable exception is the Gaussian process regressor.)
18
18
19
-
This is demonstrated in `this example <../examplex/Ensemble-regression.html>`__, where two regressors are trained on distinct subsets of the same dataset. In the figure below, the regressors are shown along with the mean predictions and the standard deviation.
19
+
This is demonstrated in `this example <../examples/Ensemble-regression.html>`__, where two regressors are trained on distinct subsets of the same dataset. In the figure below, the regressors are shown along with the mean predictions and the standard deviation.
0 commit comments