Rolling Average Over Data Stream

Problem

Given a stream of numeric data points and an integer window size k, return the rolling average after each new data point arrives once the window contains k elements. Implement an algorithm that updates the result efficiently as the window slides by one element at a time. The input is a list of numbers and an integer k; the output is a list of floating-point averages.

Constraints

• 1 <= len(nums) <= 10^5
• 1 <= k <= len(nums)
• -10^6 <= nums[i] <= 10^6
• The solution should avoid recomputing each window sum from scratch