def _sift_down(self, index):
        """ Swap with smallest child if needed
        and sift_down from that child index if swapped.
        """
        # ---start student section---
        if self._indices_of_children(index) != []:
            min_child = self._index_of_min_child(self._indices_of_children(index))
            self.comparisons += 1
            if self._data[min_child].priority < self._data[index].priority:
                self._swap(min_child, index)
                self._sift_down(min_child)
        # ===end student section===

    def _heapify(self, start_data):
        """ Turn the existing data into a heap in O(n log n) time. """
        for car in start_data:
            self.enqueue(car)

    def _fast_heapify(self, data):
        """ Turn the existing data into a heap in O(n) time. """
        # ---start student section---
        for car in data:
            assert isinstance(car, Car)
            self._data.append(car)
        for num in range((len(self._data)//self.num_children)-1,-1,-1):
            self._sift_down(num)


        # ===end student section===

    def enqueue(self, car):
        """ Add a car to the queue. """
        # We first make sure that we're only including Cars
        assert isinstance(car, Car)
        # ---start student section---
        self._data.append(car)
        self._sift_up(len(self._data)-1)
        # ===end student section===

    def dequeue(self):
        """ Take the min priority Car off the queue and return the Car object

        Must raise IndexError("Can't dequeue from empty queue!") if the queue is empty.
        """
        # ---start student section---
        if len(self._data) < 1:
            raise IndexError("Can't dequeue from empty queue!")
        dequeued_value = self._data[0]
        self._data[0] = self._data[-1]
        self._data.pop()
        self._sift_down(0)
        return dequeued_value