Classification isn't just about labels—it's about drawing lines (or curves) that separate groups. This line is called a Decision Boundary.
Your Mission
Adjust the line to separate the red points from the blue points. See what happens when the data gets messy ("Overlapping").
Class 0
Class 1
White border = correct, Black border = wrong
Data Mode
Boundary
Metrics
Accuracy:0.0%
Precision:0.0%
Recall:0.0%
TP
0
FP
0
TN
0
FN
0
What Did You Learn?
- ✓
Linear boundaries are limited: A straight line can't separate everything (like the XOR pattern).
- ✓
Real data is messy: When classes overlap, 100% accuracy is impossible.
- ✓
The boundary *is* the model: Training a model just means finding the best place to draw this line.