Data Science and Machine Learning (Theory and Projects) A to Z - RNN Architecture: Fixed Length Memory Model Exercise So

The video tutorial explains the concept of running average, which is used to compute the average of data that arrives in a stream. It starts with a basic example of calculating the average of a static set of numbers and then demonstrates how to update this average when new data points are added. The tutorial introduces a general formula for calculating the running average without recomputing the entire dataset, making it efficient for large data streams. The video concludes by mentioning a follow-up video that will relate running averages to sequence modeling.

What is the primary purpose of a running average?

• To find the maximum value in a data set

• To calculate the average of numbers in a data stream

• To compute the average of a static list of numbers

• To determine the median of a data set

How can you update the running average when a new data point is added?

• Remove the oldest data point

• Incorporate the new data point into the existing average

• Ignore the new data point

• Recompute the average from scratch

What is the advantage of using a running average?

• It is easier to understand

• It provides more accurate results

• It is faster than recalculating the entire average

• It requires less memory

Which formula is used to update the running average with a new data point?

• Add the new data point to the sum and divide by the new total number of points

• Divide the sum of all data points by the new data point

• Subtract the oldest data point and add the new one

• Multiply the current average by the number of points

What will the next video discuss in relation to running averages?

• How to find the mode

• How to compute the standard deviation

• How to model sequences

• How to calculate the median