mapclassify.gadf

mapclassify.gadf(y, method='Quantiles', maxk=15, pct=0.8)[source]

Evaluate the Goodness of Absolute Deviation Fit (GADF) of a classifier and find the minimum value of \(k\) for which gadf > pct.

Parameters:
ynumpy.array

\((n, 1)\), values to be classified.

methodstr (default ‘Quantiles’)

The classification method in: {'Quantiles', 'Fisher_Jenks', 'Maximum_Breaks', 'Natrual_Breaks'}.

maxkint (default 15)

Maximum value of \(k\) to evaluate.

pctfloat (default 0.8)

The percentage of GADF to exceed.

Returns:
kint

Number of classes.

clobject

Instance of the classifier at \(k\).

gadffloat

Goodness of absolute deviation fit (GADF).

See also

KClassifiers

Notes

The GADF is defined as:

\[GADF = 1 - \sum_c \sum_{i \in c} |y_i - y_{c,med}| / \sum_i |y_i - y_{med}|\]

where \(y_{med}\) is the global median and \(y_{c,med}\) is the median for class \(c\).

Examples

>>> import mapclassify
>>> cal = mapclassify.load_example()
>>> qgadf = mapclassify.classifiers.gadf(cal)
>>> qgadf[0]
15

>>> float(qgadf[-1])
0.3740257590909283

Quantiles fail to exceed 0.80 before 15 classes. If we lower the bar to 0.2 we see quintiles as a result

>>> qgadf2 = mapclassify.classifiers.gadf(cal, pct = 0.2)
>>> qgadf2[0]
5

>>> float(qgadf2[-1])
0.21710231966462412