What this "Missing" term means here at each node after split in Image? and also what is at leaf, is this means prediction value? I converted Output variable to 1 and 0.
I tried searching on google, but didn't get the answer. Please help
$\begingroup$
$\endgroup$
Add a comment
|
1 Answer
$\begingroup$
$\endgroup$
4
The "missing" at the different nodes are the observations for which the feature on which the split is made is missing. E.g. if the value for TotalSubmissionGuidelines is below 0.5 or missing, then the observation follows the left path of the decision tree. XGBoost supports missing values by default and learns the branch direction during training, see the documentation.
-
$\begingroup$ Will it show "missing" even I've handle the missing values in my data set? How about value on leaf node? Is it residual? $\endgroup$Pramod yadav– Pramod yadav2021-03-05 04:31:43 +00:00Commented Mar 5, 2021 at 4:31
-
$\begingroup$ I am not familiar, so I'm not sure if the missing will still show up if you have no missing values in your data, this would be something that you could easily try. With regards to the values on the leaf nodes, I would expect these to be the predicted value for all observations in that leaf node. $\endgroup$Oxbowerce– Oxbowerce2021-03-05 08:59:48 +00:00Commented Mar 5, 2021 at 8:59
-
$\begingroup$ I guess "missing" and "yes" at the node have same preference. $\endgroup$Pramod yadav– Pramod yadav2021-03-06 10:00:51 +00:00Commented Mar 6, 2021 at 10:00
-
$\begingroup$ Also with respect to values at leaf node. There are -ve values as well. $\endgroup$Pramod yadav– Pramod yadav2021-03-07 05:50:29 +00:00Commented Mar 7, 2021 at 5:50