Problem 3. [15 points] Suppose we construct a decision tree (shown below) for the data set in the following table, using the Gini index to select which features to split on. We want to predict YHeart attack Fill in the missing values, including the branches (what feature is split on, such as Xmale), the Gini index at each node, the number of samples at each node with Y = 0 and Y = 1, and the predictions at the leaf nodes. Show your calculations. Also report the training accuracy. Gini=? #Y = 0, #Y = 1 X? = 0 X = 1 Gini=? #Y = 0, #Y = 1 =? Gini=? #Y = 0, #Y = 1 Y =? 1 PATIENT ID CHEST PAIN? MALE? SMOKES? EXERCISES? HEART ATTACK? 1. yes yes no yes yes 2. yes yes yes no yes 3. no no yes no yes 4. no yes no yes no 5. yes no yes yes yes 6. no yes yes yes no