Min-max scaling uses exact feature endpoints from the displayed table.

highlighted = computed this step

Compute the small-feature range

For the small feature, min is 0, max is 4, so the range is 4.

40=44 - 0 = 4
Feature Scaling ExactlyExact min-max fractions from one displayed table.min-max scaled feature tablerowsmall rawlarge rawsmall scaledlarge scaledQ09009/10A101/40B410011small: min 0, max 4, range 4large: min 0, max 100, range 100preprocessing changes coordinate scale onlydistance and gradient-style methods see the changed coordinatesNOT training; NOT learning; NOT generalizationNOT accuracy; NOT calibration; NOT a guarantee of model qualityNOT probability truth

Compute the large-feature range

For the large feature, min is 0, max is 100, so the range is 100.

1000=100100 - 0 = 100
Feature Scaling ExactlyExact min-max fractions from one displayed table.min-max scaled feature tablerowsmall rawlarge rawsmall scaledlarge scaledQ09009/10A101/40B410011small: min 0, max 4, range 4large: min 0, max 100, range 100preprocessing changes coordinate scale onlydistance and gradient-style methods see the changed coordinatesNOT training; NOT learning; NOT generalizationNOT accuracy; NOT calibration; NOT a guarantee of model qualityNOT probability truth

Use the same rule per feature

Each displayed value uses value minus min over max minus min. The denominator is feature-specific and exact.

xscaled=xminmaxminx_{\text{scaled}}={x-\min\over \max-\min}
Feature Scaling ExactlyExact min-max fractions from one displayed table.min-max scaled feature tablerowsmall rawlarge rawsmall scaledlarge scaledQ09009/10A101/40B410011small: min 0, max 4, range 4large: min 0, max 100, range 100preprocessing changes coordinate scale onlydistance and gradient-style methods see the changed coordinatesNOT training; NOT learning; NOT generalizationNOT accuracy; NOT calibration; NOT a guarantee of model qualityNOT probability truth