Turn cumulative model output into correctly labeled category probabilities
Prevent a cumulative prediction from becoming three misleading category bars.
Read the lesson →THE CREATIVE FIELDBOOK / SNS MEDIA TECH
Practical lessons
1095 source-led reads, 6 per page. These numbered pages work without JavaScript. Search every study and lesson →
Find the best feasible bundle of indivisible projects
A high value-to-cost ratio can select the wrong bundle when projects cannot be divided.
Read the lesson →Compare complete machine assignments instead of choosing the cheapest pair
Using the cheapest machine first can force a much more expensive remaining assignment.
Read the lesson →Calculate shipment cost without losing units at transfer points
A cheap route cannot carry more units than its tightest leg permits.
Read the lesson →Calculate why equal average workloads can produce different waits
Service-time dispersion can change expected waiting even when arrival rate and mean handling time stay fixed.
Read the lesson →Check whether pairwise estimates form a usable joint matrix
A populated matrix can fail a basic variance check when its cells use different records.
Read the lesson →