Calculate Jenks Natural Breaks on complete and NaN-containing datasets, using PySAL

jenks_breaks_notnull.py

from pysal.esda.mapclassify import Natural_Breaks as nb

# Calculate Jenks natural breaks
# Assumes that there are no NaN values in Series
breaks = nb(
    df['column'].values,
    initial=300,
    k=5)

jb = pd.DataFrame({'jenks_bins': breaks.yb}, index=df['column'].index)
df = df.join(jb)

jenks_breaks_nullvalues.py

from pysal.esda.mapclassify import Natural_Breaks as nb

# Calculate Jenks natural breaks for Series which may have null values
breaks = nb(
    df['column'].dropna().values,
    initial=300,
    k=5)

jb = pd.DataFrame({'jenks_bin': breaks.yb}, index=df['column'].dropna().index)
jenks = df.join(jb)
# This still leaves us with a number of NaN values, which we have to get rid of.
# the fillna() method creates a separate class of '-1' values,
# which will end up as the lightest colour when normalising
jenks['jenks_bin'].fillna(-1, inplace=True)