ljwolf

from sklearn import metrics
from sklearn.neighbors import BallTree
import numpy, pandas
from scipy.optimize import minimize_scalar
from scipy import sparse, stats
from libpysal.weights import WSP, Kernel
from libpysal.weights.util import get_points_array

coordinates = numpy.random.normal(0, 1, size=(100, 2))

ljwolf

ljwolf

ljwolf

ljwolf

latitude,longitude,q-score,geometry
40.571148,-73.997086,10.854957,POINT (-73.997086 40.571148)
40.571148,-73.996719,13.955506,POINT (-73.996719 40.571148)
40.571159,-73.995979,11.594976,POINT (-73.99597900000001 40.571159)
40.571167,-73.995605,12.206305,POINT (-73.995605 40.571167)
40.571175,-73.998177,11.686462,POINT (-73.998177 40.571175)
40.571178,-73.994858,11.74865,POINT (-73.99485799999999 40.571178)
40.571178,-73.99781,11.360404,POINT (-73.99781 40.571178)
40.571186,-73.994476,11.910742,POINT (-73.99447600000001 40.571186)
40.571198,-73.993729,11.689811,POINT (-73.993729 40.571198)

ljwolf

import numpy
from scipy import sparse
from scipy import stats
from sklearn.base import BaseEstimator
import pysal.lib as lp


class losh(BaseEstimator):
    """Local spatial heteroscedasticity (LOSH)"""

ljwolf

import numpy as np
from scipy.spatial import distance
from scipy.special import (entr, rel_entr, kl_div)

def entropy(p, q=None):
    p = np.asarray(p)
    if p.ndim == 1:
        p = p.reshape(1,-1)
    p = p/p.sum(axis=1, keepdims=True)
    if q is None:

ljwolf

ljwolf

ljwolf

# packages in environment at /home/lw17329/anaconda/envs/ana:
#
# Name                    Version                   Build  Channel
_r-mutex                  1.0.0                     mro_2    r
affine                    2.2.1                      py_0    conda-forge
appdirs                   1.4.3            py37h28b3542_0  
asn1crypto                0.24.0                   py37_0  
atomicwrites              1.1.5                    py37_0  
atomicwrites              1.1.5                     <pip>
attrs                     18.1.0                     py_1    conda-forge