Incognition logo
Home
Learn
Practice
Leaderboard
0
VisualizationsNeural Networks & How They Learn

Binary vs Categorical Cross-Entropy

Slide the model's chance that a message is spam. With two options, one number sets both, and the loss drops as the right answer, spam, gets more likely.

Two options only
spam
75%
not spam
25%

One number decides both. If P(spam) is 75%, then P(not spam) must be 25%.

loss (truth = spam)0.29
Customize