Commit 4576cb4a authored by Gregory Herschlag's avatar Gregory Herschlag
Browse files

adding wrapper for fuzziness

parent 62c05dd8

src/countyClusterer/cluster/clusters_full3.py

+8 −4
Original line number Diff line number Diff line
@@ -190,7 +190,9 @@ def clusters(graph, pops, ndistricts, tolerance, verbose=0, fuzz=0): 
    memo = {} #memoization

    outout = set()

    while result[-1] and max(len(x[0]) for x in result[-1]) > 0: #while there are counties left

        print(len(outout), len(result), len(result[-1]), min(len(x[0]) for x in result[-1]), max(len(x[0]) for x in result[-1])) #cluster size, number of options, counties left

        if verbose > 0:

            print(len(outout), len(result), len(result[-1]), 

                  min(len(x[0]) for x in result[-1]), max(len(x[0]) for x in result[-1])) #cluster size, number of options, counties left

        adding = set()

        j = 0

        cluster_size = len(result) + 1
@@ -199,7 +201,7 @@ def clusters(graph, pops, ndistricts, tolerance, verbose=0, fuzz=0): 
                outout.add(clusters)

                continue

            j += 1

            if j % 200 == 0 and adding:

            if j % 200 == 0 and adding and verbose > 0:

                print(j, best_so_far, min(len(x[0]) for x in adding), max(len(x[0]) for x in adding), len(adding))

            ndistricts_left = ndistricts

            for (cluster, districts) in clusters:
@@ -229,8 +231,10 @@ def clusters(graph, pops, ndistricts, tolerance, verbose=0, fuzz=0): 
    if not outout:

        outout.update({x[1] | frozenset({frozenset(x[0])}) for x in result[-2]})

    M = max(len(x) for x in outout)

    if verbose > 0:

        print(M)

    trueout = [x for x in outout if len(x) == M]

    if verbose > 0

        print(len(trueout))

    return trueout

src/countyClusterer/qgCountyClusters.py

+27 −21
Original line number Diff line number Diff line 
import geopandas as gpd

import os

import pandas as pd

import random



import clusters

import clusters_full3

import qgGeoDFDataExtraction as gpdExtractor



from importlib import reload

reload(clusters)

reload(clusters_full3)



###############################################################################



################################################################################



def clusterCountiesDF(gpdFile, districts, cntyField="", popField="",

                      adjacency="Rook", adjFldName="Neighbors", 

                      adjDelim=", ", tol=0.05, verbose=False, 

                      verboseLevel = 0):

                      verboseLevel=0, fuzz=0):

    

    adjacencyDict = gpdExtractor.adjacencies(gpdFile, adjacency, cntyField, 

                                             adjFldName, adjDelim, verbose)
@@ -22,17 +23,22 @@ def clusterCountiesDF(gpdFile, districts, cntyField = "", popField = "", 
        raise Exception("Empty population field")

    popDict = gpdExtractor.extractDataDict(gpdFile, cntyField, popField)

    cntyClusterings = clusterCounties(adjacencyDict, popDict, districts, tol, 

                                   verbose, verboseLevel)

                                      verbose, verboseLevel, fuzz)

    return cntyClusterings



################################################################################

###############################################################################





def clusterCounties(cntyAdjacency, popData, districts, tol=0.05, 

                    verbose = False, verboseLevel = 1):

    if not verbose:

        verboseLevel = 0

    cntyClusterings = clusters.clusters(cntyAdjacency, popData, districts, tol, 

                                     verbose)

                    verbose=False, verboseLevel=1, fuzz=0):

    # if not verbose:

    #     verboseLevel = 0

    if fuzz == 0:

        cntyClusterings = clusters.clusters(cntyAdjacency, popData, districts, 

                                            tol, verbose)

    else:

        cntyClusterings = clusters_full3.clusters(cntyAdjacency, popData, 

                                                  districts, tol, verbose, fuzz)

    return cntyClusterings



################################################################################