Predicting Cryoxide Can Sales
From Cyberchase
00:00 - 03:11
3m 11s

The CyberSquad ordered too many Cryoxide cans for the shop, and they try to understand where they went wrong. After they obtain historical sales data, they identify patterns in the data and predict the number of cans that they need to order each day of the week to prevent having a surplus.

Comments

Please sign in to write a comment.

Related Clips

Slider finds Cyroxide receipts for the month of July from the previous year, and the CyberSquad writes down on a calendar how many Cyroxide cans were sold on each day. Excluding an outlier, a holiday, they find a pattern for each week: an average of 30 cans are sold on weekdays, and an average of 60 cans are sold on weekends. They use the pattern they've identified to accurately place an order for the next six days of business.
Charlie realizes that the reason he is having trouble finding a suspect is because he is getting bad information. He explains how bad or corrupt data can skew the results and make it impossible to find a correct formula to find a suspect
Multiplying a number by ten will move the first digit to the left by 1.
Uses examples of sequential music and text generation to show how and when recurrent neural networks (specifically, long short-term memory networks) are useful.
In Pacman-style movement, Mathman must eat all the numbers that make the proposed inequality true; otherwise, he will be eaten by Mr. Glitch. After Mathman assesses each number, the number is plugged into the equation to illustrate to the viewer that it is either correct or incorrect.