topic Re: Decision Tree Interpretation in Using Dataiku DSS
https://community.dataiku.com/t5/Using-Dataiku-DSS/Decision-Tree-Interpretation/m-p/15203#M6454
<P>Hi <LI-USER uid="1594"></LI-USER> <BR /><BR />Could you detail the explanation a bit more?<BR /><BR />To get a probability of a class I would have to give an input X that contains values for the features (I assume you use a predict_proba(X) method on a DecisionTreeClassifier from sklearn under the hood). When you are on a tree node, which values/inputs are used to calculate this class probability? Are these all the datapoints that the node contains and its averaged prediction probabilities, or some averaged values for X based on the datapoints that gives one prediction class probability?<BR /><BR /><BR /></P>Thu, 25 Mar 2021 13:36:45 GMTpvannies2021-03-25T13:36:45ZDecision Tree Interpretation
https://community.dataiku.com/t5/Using-Dataiku-DSS/Decision-Tree-Interpretation/m-p/14878#M6322
<P>I created a Decision Tree predictive model, and I was wondering if you could help me understand the difference between the %'s in Probabilities and Target Classes when I view the decision tree itself. What do each of these %'s represent? Below is a screen shot. </P><P> </P><P><span class="lia-inline-image-display-wrapper lia-image-align-inline" image-alt="jm596353_0-1615490324263.jpeg" style="width: 400px;"><img src="https://community.dataiku.com/t5/image/serverpage/image-id/2952i3C4C07358C2A33F4/image-size/medium?v=v2&px=400" role="button" title="jm596353_0-1615490324263.jpeg" alt="jm596353_0-1615490324263.jpeg" /></span></P><P> </P>Thu, 11 Mar 2021 19:19:03 GMThttps://community.dataiku.com/t5/Using-Dataiku-DSS/Decision-Tree-Interpretation/m-p/14878#M6322jm5963532021-03-11T19:19:03ZRe: Decision Tree Interpretation
https://community.dataiku.com/t5/Using-Dataiku-DSS/Decision-Tree-Interpretation/m-p/14880#M6323
<P>Hi,</P>
<P><SPAN>Probabilities are the probabilities of each class as predicted by tree, whereas t</SPAN><SPAN>arget classes is the distribution of data in the training set corresponding to the given tree node.</SPAN></P>
<P>Hope this helps!</P>
<P>Kim</P>Thu, 11 Mar 2021 20:25:25 GMThttps://community.dataiku.com/t5/Using-Dataiku-DSS/Decision-Tree-Interpretation/m-p/14880#M6323KimmyC2021-03-11T20:25:25ZRe: Decision Tree Interpretation
https://community.dataiku.com/t5/Using-Dataiku-DSS/Decision-Tree-Interpretation/m-p/15203#M6454
<P>Hi <LI-USER uid="1594"></LI-USER> <BR /><BR />Could you detail the explanation a bit more?<BR /><BR />To get a probability of a class I would have to give an input X that contains values for the features (I assume you use a predict_proba(X) method on a DecisionTreeClassifier from sklearn under the hood). When you are on a tree node, which values/inputs are used to calculate this class probability? Are these all the datapoints that the node contains and its averaged prediction probabilities, or some averaged values for X based on the datapoints that gives one prediction class probability?<BR /><BR /><BR /></P>Thu, 25 Mar 2021 13:36:45 GMThttps://community.dataiku.com/t5/Using-Dataiku-DSS/Decision-Tree-Interpretation/m-p/15203#M6454pvannies2021-03-25T13:36:45ZRe: Decision Tree Interpretation
https://community.dataiku.com/t5/Using-Dataiku-DSS/Decision-Tree-Interpretation/m-p/15220#M6459
<P>Hi, <BR /><BR /><SPAN>So the probabilities that you see under TARGET CLASSES are derived from the proportion of samples in the node that belong to each class.<BR /><BR /></SPAN><SPAN>The probabilities under PROBABILITIES are what the model would predict if the node was final (i.e, a leaf). All the observations falling into that node would receive the same probability prediction so there is no need to take any average.<BR /><BR />I hope this helps!<BR /><BR />Best, <BR /><BR />Regards<BR /><BR />Jean-Yves<BR /><BR /><BR /></SPAN></P>Thu, 25 Mar 2021 21:09:09 GMThttps://community.dataiku.com/t5/Using-Dataiku-DSS/Decision-Tree-Interpretation/m-p/15220#M6459Jean-Yves2021-03-25T21:09:09Z