def convert_opt_data_to_gerrychain_input(state, opt_data='', plan=None):
state_df, adj_graph, _, _ = load_opt_data(state, opt_data)
election_df = load_election_df(state, opt_data)
state_df = state_df[['population']]
metric_df = pd.concat([state_df, election_df], axis=1)
if plan is not None:
plan_inverse_map = {}
for district_ix, tract_ixs in plan.items():
for tix in tract_ixs:
plan_inverse_map[tix] = district_ix
metric_df['initial_plan'] = pd.Series(plan_inverse_map, dtype=str)
else:
k = constants.seats[state]['house']
ideal_pop = state_df.population.sum() / k
nx.set_node_attributes(adj_graph, metric_df.T.to_dict())
plan = recursive_tree_part(adj_graph, range(k), ideal_pop,
'population', 0.01)
metric_df['initial_plan'] = pd.Series(plan, dtype=str)
nx.set_node_attributes(adj_graph, metric_df.T.to_dict())
elections = [
Election(election, {
'Democratic': 'D_' + election,
'Republican': 'R_' + election
})
for election in wrappers[state]().election_columns(include_party=False)
]
return adj_graph, elections
def updaters_MC(settings):
my_updaters = {
"cut_edges": cut_edges,
"population": updaters.Tally("TOTPOP", alias = "population"),
"avg_pop_dist": avg_pop_dist,
"pop_dist_pct" : pop_dist_pct,
"area_land": updaters.Tally("ALAND10", alias = "area_land"),
"area_water": updaters.Tally("AWATER10", alias = "area_water"),
"Perimeter": updaters.Tally("perimeter", alias = "Perimeter"),
"Area": updaters.Tally("area", alias = "Area")}
num_elections = settings['num_elections']
election_names = settings['election_names']
election_columns = settings['election_columns']
num_steps = settings['num_steps']
interval = settings['interval']
pop_tol = settings['pop_tol']
elections = [
Election(
election_names[i],
{"Democratic": election_columns[i][0], "Republican": election_columns[i][1]},
) for i in range(num_elections)]
election_updaters = {election.name: election for election in elections}
my_updaters.update(election_updaters)
return(my_updaters)
def run_simple2(graph):
election = Election("2014 Senate", {
"Democratic": "sen_blue",
"Republican": "sen_red"
},
alias="2014_Senate")
initial_partition = Partition(graph,
assignment="con_distri",
updaters={
"2014_Senate":
election,
"population":
Tally("population", alias="population"),
"exterior_boundaries":
exterior_boundaries,
"interior_boundaries":
interior_boundaries,
"perimeter":
perimeter,
"boundary_nodes":
boundary_nodes,
"cut_edges":
cut_edges,
"area":
Tally("area", alias="area"),
"cut_edges_by_part":
cut_edges_by_part,
"county_split":
county_splits('county_split',
"COUNTY_ID"),
"black_population":
Tally("black_pop",
alias="black_population"),
})
districts_within_tolerance_2 = lambda part: districts_within_tolerance(
part, 'population', 0.3)
is_valid = Validator(
[single_flip_contiguous, districts_within_tolerance_2])
chain = MarkovChain(
proposal=propose_random_flip,
is_valid=is_valid,
accept=always_accept,
# accept=accept, #THe acceptance criteria is what needs to be defined ourselves - to match the paper
initial_state=initial_partition,
total_steps=30,
)
efficiency_gaps = []
wins = []
for partition in chain:
if (hasattr(partition, 'accepted') and partition.accepted):
efficiency_gaps.append(
gerrychain.scores.efficiency_gap(partition["2014_Senate"]))
wins.append(partition["2014_Senate"].wins("Democratic"))
return (efficiency_gaps, wins, partition)
"CVAP": updaters.Tally(CVAP, alias = "CVAP"),
"WCVAP": updaters.Tally(WCVAP, alias = "WCVAP"),
"HCVAP": updaters.Tally(HCVAP, alias = "HCVAP"),
"BCVAP": updaters.Tally(BCVAP, alias = "BCVAP"),
"Sum_CX": updaters.Tally(C_X, alias = "Sum_CX"),
"Sum_CY": updaters.Tally(C_Y, alias = "Sum_CY"),
"cut_edges": cut_edges,
"demo_percents": demo_percents,
"final_elec_model": final_elec_model,
"num_county_splits": num_county_splits,
"num_cut_edges": num_cut_edges
}
#updater functions
elections_track = [
Election("PRES16", {"Democratic": 'ClintonD_16G_President' , "Republican": 'TrumpR_16G_President'}, alias = "PRES16"),
Election("SEN16", {"Democratic": 'CampbellD_16G_US_Sen' , "Republican": 'KennedyR_16G_US_Sen'}, alias = "SEN16"),
]
election_updaters = {election.name: election for election in elections_track}
my_updaters.update(election_updaters)
election_functions = [Election(j, candidates[j]) for j in elections]
election_updaters = {election.name: election for election in election_functions}
my_updaters.update(election_updaters)
#initial partition########################
total_population = state_gdf[tot_pop].sum()
ideal_population = total_population/num_districts
if start_map == 'new_seed':
start_map = recursive_tree_part(graph, range(num_districts), ideal_population, tot_pop, pop_tol, 3)
#county_col = "COUNTYFP10"
pop_col = "TOTPOP"
df["CPOP"] = df["TOTPOP"] - df["NCPOP"]
ccol = "CPOP"
uid = "ID"
num_districts = 14
graph = Graph.from_geodataframe(df, ignore_errors=True)
graph.add_data(df, list(df))
graph = nx.relabel_nodes(graph, df[uid])
elections = [
Election("PRES16", {
"Democratic": "PRES16D",
"Republican": "PRES16R"
}),
Election("SEN16", {
"Democratic": "SEN16D",
"Republican": "SEN16R"
})
]
#my_updaters = {"population" : updaters.Tally("TOTPOP", alias="population")}
my_updaters = {
"population": Tally(pop_col, alias="population"),
"cpop": Tally(ccol, alias="cpop"),
"cut_edges": cut_edges
}
election_updaters = {election.name: election for election in elections}
my_updaters.update(election_updaters)
def main(graph_json, shp, n_steps, output_dir, prefix, seed, pop_col, pop_tol,
plan_col, reproject, election):
os.makedirs(output_dir, exist_ok=True)
has_geometry = False
if not shp and not graph_json:
print('Specify a shapefile or a NetworkX-format graph '
'JSON file.',
file=sys.stderr)
sys.exit(1)
elif shp and not graph_json:
gdf = gpd.read_file(shp)
if reproject:
gdf = reprojected(gdf)
graph = Graph.from_geodataframe(gdf)
has_geometry = True
elif graph_json and not shp:
graph = Graph.from_json(graph_json)
else:
graph = Graph.from_json(graph_json)
gdf = gpd.read_file(shp)
if reproject:
gdf = reprojected(gdf)
print('Appending geometries from shapefile to graph...')
graph.geometry = gdf.geometry # TODO: is this always valid?
has_geometry = True
my_updaters = {'population': updaters.Tally(pop_col, alias='population')}
if election:
election_up = Election('election', {
'Democratic': election[0],
'Republican': election[1]
})
my_updaters['election'] = election_up
initial_state = GeographicPartition(graph,
assignment=plan_col,
updaters=my_updaters)
normal_chain = RecomChain(graph=graph,
total_steps=n_steps,
initial_state=initial_state,
pop_col=pop_col,
pop_tol=pop_tol,
reversible=False,
seed=seed)
reversible_chain = RecomChain(graph=graph,
total_steps=n_steps,
initial_state=initial_state,
pop_col=pop_col,
pop_tol=pop_tol,
reversible=True,
seed=seed)
normal_plans = [plan for plan in tqdm(normal_chain)]
reversible_plans = [plan for plan in tqdm(reversible_chain)]
cut_edges_fig(output_dir, prefix, normal_plans, reversible_plans)
longest_boundary_fig(output_dir, prefix, normal_plans, reversible_plans)
if has_geometry:
demo_plans(output_dir, '_'.join([prefix, 'recom']), normal_plans,
n_steps, n_steps // 25)
demo_plans(output_dir, '_'.join([prefix, 'reversible_recom']),
reversible_plans, n_steps, n_steps // 25)
if election:
election_hists(output_dir, 'dem_vote_share', 'election', 'Democratic',
normal_plans, reversible_plans)
acceptance_stats(output_dir, '_'.join([prefix, 'recom']), normal_plans)
acceptance_stats(output_dir, '_'.join([prefix, 'reversible_recom']),
reversible_plans)
"BVAP": updaters.Tally(BVAP_col, alias="BVAP"),
"HVAP": updaters.Tally(HVAP_col, alias="HVAP"),
"AVAP": updaters.Tally(AVAP_col, alias="AVAP"),
"NVAP": updaters.Tally(Native_col, alias="NVAP"),
"PVAP": updaters.Tally(Pacific_col, alias="PVAP"),
"VAP": updaters.Tally(VAP_col, alias="VAP"),
"county_splits": county_splits("county_splits", county_assignment_col)
}
num_elections = len(election_names)
elections = [
Election(
election_names[i],
{
"Dem": election_columns[i][0],
"Rep": election_columns[i][1]
},
) for i in range(num_elections)
]
election_updaters = {election.name: election for election in elections}
updaters.update(election_updaters)
# ---- Proposed Partition -----
initial_partition = GeographicPartition(graph, district_assignment_col,
updaters)
# ---- Chain Run ------
import numpy as np
import seaborn as sns
from FKT import FKT
from gerrychain import (Election, Graph, MarkovChain,
Partition, accept, constraints, updaters)
from gerrychain.proposals import recom
from gerrychain.updaters import cut_edges
whole_start = time.time()
# Initialization
df = gpd.read_file("./AK_precincts_ns/AK_precincts_ns/AK_precincts_ns.shp")
df["nAMIN"] = df["TOTPOP"] - df["AMIN"]
elections = [
Election("GOV18x", {"Democratic": "GOV18D_x", "Republican": "GOV18R_x"}),
Election("USH18x", {"Democratic": "USH18D_x", "Republican": "USH18R_x"}),
Election("GOV18ns", {"Democratic": "GOV18D_NS", "Republican": "GOV18R_NS"}),
Election("USH18ns", {"Democratic": "USH18D_NS", "Republican": "USH18R_NS"}),
Election("Native_percent", {"Native": "AMIN", "nonNative": "nAMIN"}),
]
my_updaters = {
"population": updaters.Tally("POPULATION", alias="population"),
"cut_edges": cut_edges,
}
election_updaters = {election.name: election for election in elections}
my_updaters.update(election_updaters)
def chain(iterations):
idef = random.randint(1, 10000)
graph = Graph.from_json("./PA_VTD.json")
election = Election("SEN12", {"Dem": "USS12D", "Rep": "USS12R"})
initial_partition = GeographicPartition(graph,
assignment="2011_PLA_1",
updaters={
"cut_edges":
cut_edges,
"population":
Tally("TOT_POP",
alias="population"),
"SEN12":
election
})
ideal_population = sum(
initial_partition["population"].values()) / len(initial_partition)
# We use functools.partial to bind the extra parameters (pop_col, pop_target, epsilon, node_repeats)
# of the recom proposal.
proposal = partial(recom,
pop_col="TOT_POP",
pop_target=ideal_population,
epsilon=0.02,
node_repeats=2)
chain = MarkovChain(proposal=proposal,
constraints=[republican_constraint],
accept=contiguous,
initial_state=initial_partition,
total_steps=iterations + 100)
count = 0
metrics = []
boundary_nodes = []
boundary_weighted = []
for partition in chain.with_progress_bar():
mm = mean_median(partition["SEN12"])
p = pp(partition)
bias = partisan_bias(partition["SEN12"])
gini = partisan_gini(partition["SEN12"])
gap = efficiency_gap(partition["SEN12"])
cut = len(partition["cut_edges"])
if count >= 100:
metrics.append((mm, p, bias, gini, gap, cut))
nodes = [0] * 8921
bnodes = [0] * 8921
for edge in partition["cut_edges"]:
nodes[edge[0]] = 1
nodes[edge[1]] = 1
bnodes[edge[0]] += 1
bnodes[edge[1]] += 1
boundary_nodes.append(nodes)
boundary_weighted.append(bnodes)
if count % 100 == 0:
print(idef, count, mm, p, bias, gini, gap, cut,
partition["SEN12"].wins("Rep"))
count += 1
return metrics, boundary_nodes, boundary_weighted
c for c in components if len(c) != biggest_component_size
]
problem_nodes = [
node for component in problem_components for node in component
]
problem_geoids = [graph.nodes[node]["GEOID10"] for node in problem_nodes]
is_a_problem = df["GEOID10"].isin(problem_geoids)
largest_component_size = max(len(c) for c in components)
to_delete = [c for c in components if len(c) != largest_component_size]
for c in to_delete:
for node in c:
graph.remove_node(node)
election = Election("PRETOT16", {"Dem": "PREDEM16", "Rep": "PREREP16"})
#Create initial parition based on congressional districts
initial_partition = Partition(graph,
assignment="CON",
updaters={
"cut_edges":
cut_edges,
"population":
Tally("PERSONS", alias="population"),
"PRETOT16":
election
})
my_updaters = {
"population": Tally("PERSONS", alias="population"),
"PRETOT16": election
import matplotlib.pyplot as plt
from gerrychain import (Election, GeographicPartition, Graph, MarkovChain,
Partition, accept, constraints, proposals, updaters)
from gerrychain.tree_proposals import recom
# ## Setting up the initial districting plan
# Load the graph we want to use.
graph = Graph.from_json("./graphs/rook/PA_VTD.json")
# Configure our elections, telling GerryChain which column names for our shapefiles
# correspond to vote totals.
elections = [
Election("SEN10", {
"Democratic": "SEN10D",
"Republican": "SEN10R"
}),
Election("SEN12", {
"Democratic": "USS12D",
"Republican": "USS12R"
}),
Election("SEN16", {
"Democratic": "T16SEND",
"Republican": "T16SENR"
}),
Election("PRES12", {
"Democratic": "PRES12D",
"Republican": "PRES12R"
}),
Election("PRES16", {
"Democratic": "T16PRESD",
"BCVAP": updaters.Tally(BCVAP, alias="BCVAP"),
"Sum_CX": updaters.Tally(C_X, alias="Sum_CX"),
"Sum_CY": updaters.Tally(C_Y, alias="Sum_CY"),
"cut_edges": cut_edges,
"num_cut_edges": num_cut_edges,
"num_county_splits": num_county_splits,
"demo_percents": demo_percents,
"final_elec_model": final_elec_model,
"centroids": centroids
}
#add elections updaters
elections_track = [
Election("PRES16", {
"Democratic": 'ClintonD_16G_President',
"Republican": 'TrumpR_16G_President'
},
alias="PRES16"),
Election("PRES12", {
"Democratic": 'ObamaD_12G_President',
"Republican": 'RomneyR_12G_President'
},
alias="PRES12"),
Election("SEN18", {
"Democratic": "ORourkeD_18G_US_Sen",
"Republican": 'CruzR_18G_US_Sen'
},
alias="SEN18"),
Election("GOV18", {
"Democratic": "ValdezD_18G_Governor",
"Republican": 'AbbottR_18G_Governor'
df["nBCVAP"] = df["CVAP"] - df["BCVAP"]
df["BHCPOP"] = pd.to_numeric(df["BCPOP"] + df[HCPOP_cols].sum(axis=1))
df["nBHCPOP"] = df["CPOP"] - df["BHCPOP"]
df["BHVAP"] = pd.to_numeric(df["BVAP"] + df[HVAP_cols].sum(axis=1))
df["nBHVAP"] = df["VAP"] - df["BHVAP"]
df["BHCVAP"] = pd.to_numeric(df["BCVAP"] + df[HCVAP_cols].sum(axis=1))
df["nBHCVAP"] = df["CVAP"] - df["BHCVAP"]
graph.add_data(df, columns=new_cols)
elections = [
Election("BPOP", {
"BPOP": "BPOP",
"nBPOP": "nBPOP"
}),
Election("BCPOP", {
"BCPOP": "BCPOP",
"nBCPOP": "nBCPOP"
}),
Election("BVAP", {
"BVAP": "BVAP",
"nBVAP": "nBVAP"
}),
Election("BCVAP", {
"BCVAP": "BCVAP",
"nBCVAP": "nBCVAP"
}),
Election("BHPOP", {
"BHPOP": "BHPOP",
["W101216D", "W101216R"],
["W1216D", "W1216R"],
]
graph = Graph.from_json(graph_path)
updaters = {
"population": updaters.Tally("TOT_POP", alias="population"),
"cut_edges": cut_edges,
}
elections = [
Election(
election_names[i],
{"Democratic": election_columns[i][0], "Republican": election_columns[i][1]},
)
for i in range(num_elections)
]
election_updaters = {election.name: election for election in elections}
updaters.update(election_updaters)
initial_partition = Partition(graph, "2011_PLA_1", updaters)
ideal_population = sum(initial_partition["population"].values()) / len(
initial_partition
)
ITERS = args.n
EPS = epsilons[args.map]
ELECTS = elections_by_year[args.year]
DEMO_COLS = ["TOTPOP", "VAP", "CPOP", "CVAP",
"BPOP", "HPOP", "WPOP", "BPOP_perc", "HPOP_perc", "WPOP_perc",
"BCPOP", "HCPOP", "WCPOP", "BCPOP_perc", "HCPOP_perc", "WCPOP_perc",
"BCVAP", "HCVAP", "WCVAP", "BCVAP_perc", "HCVAP_perc", "WCVAP_perc"]
## Pull in graph and set up updaters
print("Reading in Data/Graph")
with open("../graphs/GA_precincts_{}_graph.p".format(args.year), "rb") as f_in:
graph = pickle.load(f_in)
elections = [Election(e, {"Dem": "{}D".format(e), "Rep": "{}R".format(e)}) for e in ELECTS]
ga_updaters = {"population" : Tally(POP_COL, alias="population"),
"cut_edges": cut_edges,
"TOTPOP": Tally("TOTPOP18", alias="TOTPOP"),
"VAP": Tally("VAP18", alias="VAP"),
"CPOP": Tally("CPOP18", alias="CPOP"),
"CVAP": Tally("CVAP18", alias="CVAP"),
"BPOP": Tally("NH_BLACK18", alias="BPOP"),
"HPOP": Tally("HISP18", alias="HPOP"),
"WPOP": Tally("NH_WHITE18", alias="WPOP"),
"BCPOP": Tally("BCPOP18", alias="BCPOP"),
"HCPOP": Tally("HCPOP18", alias="HCPOP"),
"WCPOP": Tally("WCPOP18", alias="WCPOP"),
"BCVAP": Tally("BCVAP18", alias="BCVAP"),
def chain(iterations):
idef = random.randint(1, 10000)
graph = Graph.from_json("../data/PA_init/PA_VTD.json")
election = Election("SEN12", {"Dem": "USS12D", "Rep": "USS12R"})
initial_partition = GeographicPartition(
graph,
assignment="2011_PLA_1",
updaters={
"cut_edges": cut_edges,
"population": Tally("TOT_POP", alias="population"),
"SEN12": election
}
)
ideal_population = sum(initial_partition["population"].values()) / len(initial_partition)
# We use functools.partial to bind the extra parameters (pop_col, pop_target, epsilon, node_repeats)
# of the recom proposal.
proposal = partial(recom,
pop_col="TOT_POP",
pop_target=ideal_population,
epsilon=0.02,
node_repeats=2
)
chain = MarkovChain(
proposal=proposal,
constraints=[],
accept=contiguous,
initial_state=initial_partition,
total_steps=85*iterations + 17000
)
count = 0
metrics = []
boundary_nodes = []
boundary_weighted = []
for partition in chain.with_progress_bar():
mm = mean_median(partition["SEN12"])
p = pp(partition)
bias = partisan_bias(partition["SEN12"])
gini = partisan_gini(partition["SEN12"])
gap = efficiency_gap(partition["SEN12"])
cut = len(partition["cut_edges"])
if count >= 17000:
if count % 85 == 0:
metrics.append((mm, p, bias, gini, gap, cut, partition["SEN12"].wins("Rep")))
nodes = [0]*8921
bnodes = [0]*8921
for edge in partition["cut_edges"]:
nodes[edge[0]] = 1
nodes[edge[1]] = 1
bnodes[edge[0]] += 1
bnodes[edge[1]] += 1
boundary_nodes.append(nodes)
boundary_weighted.append(bnodes)
assign = {i: partition.assignment[i] for i in partition.assignment}
shape["CD"] = shape.index.map(assign)
this_map = shape.dissolve(by='CD')
this_map.plot(color='white', edgecolor='black')
plt.axis('off')
fig = plt.gcf()
fig.set_size_inches((15,9), forward=False)
fig.savefig("../images/PA_neutral/" + str(idef) + str(int((count-17000)/85)+1) + ".png", dpi=600, bbox_inches="tight", pad_inches=0)
plt.close()
if count % 8500 == 0:
print(idef, count, mm, p, bias, gini, gap, cut, partition["SEN12"].wins("Rep"))
else:
if count%1000 == 0:
print(idef, "Mixing...... Iteration", count, "/17000")
count += 1
return metrics, boundary_nodes, boundary_weighted, idef
# In[57]:
path_to_pa = r'C:\Users\adika\Downloads\PA_VTDs.json'
# In[58]:
from gerrychain import Graph, Partition, Election
from gerrychain.updaters import Tally, cut_edges
graph = Graph.from_json(path_to_pa)
election = Election("SEN12", {"Dem": "USS12D", "Rep": "USS12R"})
initial_partition = Partition(
graph,
assignment="CD_2011",
updaters={
"cut_edges": cut_edges,
"population": Tally("TOTPOP", alias="population"),
"SEN12": election
}
)
# In[59]:
totpop[i] += graph.node[n]["TOTPOP"]
graph.node[n]["nBVAP"] = graph.node[n][
vap_list[i]] - graph.node[n]["BVAP"]
starts = []
for i in range(4):
starts.append(
recursive_tree_part(graph_list[i], range(num_districts),
totpop[i] / num_districts, "TOTPOP", .02, 1))
updater = {
"population": updaters.Tally("TOTPOP", alias="population"),
"cut_edges": cut_edges,
"BVAP": Election("BVAP", {
"BVAP": "BVAP",
"nBVAP": "nBVAP"
})
}
initial_partitions = []
proposals = []
compactness_bounds = []
chains = []
for i in range(4):
initial_partitions.append(Partition(graph_list[i], starts[i], updater))
proposals.append(
partial(recom,
pop_col="TOTPOP",
pop_target=totpop[i] / num_districts,
def run_simple(graph, num_samples = 80000, wp = 3000, wi = 2.5, wc = 0.4, wm = 800, get_parts = 5):
election = Election(
"2014 Senate",
{"Democratic": "sen_blue", "Republican": "sen_red"},
alias="2014_Senate"
)
initial_partition = Partition(
graph,
assignment="2014_CD",
updaters={
"2014_Senate": election,
"population": Tally("population", alias="population"),
"exterior_boundaries": exterior_boundaries,
"interior_boundaries": interior_boundaries,
"perimeter": perimeter,
"boundary_nodes": boundary_nodes,
"cut_edges": cut_edges,
"area": Tally("area", alias="area"),
"cut_edges_by_part": cut_edges_by_part,
"county_split" : county_splits( 'county_split', "COUNTY_ID"),
"black_population": Tally("black_pop", alias = "black_population"),
}
)
def vra_validator(part):
if(vra_district_requirement(part, 2, [0.4, 0.335]) > 0):
return False
return True
districts_within_tolerance_2 = lambda part : districts_within_tolerance(part, 'population', 0.12)
is_valid = Validator([single_flip_contiguous, districts_within_tolerance_2, vra_validator ])
def accept(partition, counter):
if(counter < 40000):
beta = 0
elif(counter < 60000):
beta = (counter - 40000) /(60000-40000)
else:
beta = 1
return(metro_scoring_prob(partition, beta, wp, wi, wc, wm))
chain = MarkovChainAnneal(
proposal=propose_random_flip,
is_valid=is_valid,
accept=accept,
initial_state=initial_partition,
total_steps= num_samples * 100,
)
# going to show 5 partitions from this sample
part_con = max(1, num_samples / get_parts)
efficiency_gaps = []
wins = []
voting_percents = []
sample_parts = []
tbefore = time.time()
for partition in chain:
if(hasattr(partition, 'accepted') and partition.accepted):
efficiency_gaps.append(gerrychain.scores.efficiency_gap(partition["2014_Senate"]))
wins.append(partition["2014_Senate"].wins("Democratic"))
voting_percents.append(partition["2014_Senate"].percents("Democratic"))
num_s = len(wins)
if(num_s % part_con) == 0:
sample_parts.append(partition)
if(num_s> num_samples):
break
if(num_s % 1000 == 0):
tnext = time.time()
print("It took %i time to get 1000 samples" % (tnext - tbefore))
tbefore = time.time()
return(efficiency_gaps, wins, voting_percents, sample_parts)
def run_simple(graph):
election = Election("2016 President", {
"Democratic": "T16PRESD",
"Republican": "T16PRESR"
},
alias="2016_President")
initial_partition = Partition(graph,
assignment="2011_PLA_1",
updaters={
"2016_President":
election,
"population":
Tally("TOT_POP", alias="population"),
"perimeter":
perimeter,
"exterior_boundaries":
exterior_boundaries,
"interior_boundaries":
interior_boundaries,
"boundary_nodes":
boundary_nodes,
"cut_edges":
cut_edges,
"area":
Tally("area", alias="area"),
"cut_edges_by_part":
cut_edges_by_part,
"county_split":
county_splits('county_split',
"COUNTYFP10"),
"black_population":
Tally("BLACK_POP",
alias="black_population"),
})
efficiency_gaps = []
wins = []
beta = 0
wp = 1
wi = 1
wc = 0
wm = 1
def accept(partition):
return (metro_scoring_prob(partition, beta, wp, wi, wc, wm))
is_valid = Validator([single_flip_contiguous, districts_within_tolerance])
chain = MarkovChain(
proposal=propose_random_flip,
is_valid=is_valid,
#accept = always_accept,
accept=
accept, #THe acceptance criteria is what needs to be defined ourselves - to match the paper
initial_state=initial_partition,
total_steps=30)
for partition in chain:
if (hasattr(partition, 'accepted') and partition.accepted):
efficiency_gaps.append(
gerrychain.scores.efficiency_gap(partition["2016_President"]))
wins.append(partition["2016_President"].wins("Democratic"))
return (efficiency_gaps, wins, partition)
def MC_sample(jgraph, settings, save_part = True):
"""
:param jgraph: gerrychain Graph object
:param settings: settings dictionary (possibly loaded from a yaml file) with election info, MC parameters, and constraints params (see settings.yaml file for an example of the structure needed)
:param save_part: True is you want to save the partition as json
:returns: a list of partitions sapmpled every interval step
"""
my_updaters = {
"cut_edges": cut_edges,
"population": updaters.Tally("TOTPOP", alias = "population"),
"avg_pop_dist": avg_pop_dist,
"pop_dist_pct" : pop_dist_pct,
"area_land": updaters.Tally("ALAND10", alias = "area_land"),
"area_water": updaters.Tally("AWATER10", alias = "area_water"),
"Perimeter": updaters.Tally("perimeter", alias = "Perimeter"),
"Area": updaters.Tally("area", alias = "Area")
}
num_elections = settings['num_elections']
election_names = settings['election_names']
election_columns = settings['election_columns']
num_steps = settings['num_steps']
interval = settings['interval']
pop_tol = settings['pop_tol']
MC_type = settings['MC_type']
elections = [
Election(
election_names[i],
{"Democratic": election_columns[i][0], "Republican": election_columns[i][1]},
)
for i in range(num_elections)
]
election_updaters = {election.name: election for election in elections}
my_updaters.update(election_updaters)
initial_partition = Partition(jgraph, "CD", my_updaters) # by typing in "CD," we are saying to put every county into the congressional district that they belong to
print('computed initial partition')
ideal_population = sum(initial_partition["population"].values()) / len(
initial_partition
)
pop_constraint = constraints.within_percent_of_ideal_population(initial_partition, pop_tol)
compactness_bound = constraints.UpperBound(
lambda p: len(p["cut_edges"]), 2 * len(initial_partition["cut_edges"])
)
proposal = partial(
recom, pop_col=pop_col, pop_target=ideal_population, epsilon=pop_tol, node_repeats=1)
constraints_=[pop_constraint, compactness_bound]
if MC_type == "flip":
proposal = propose_random_flip
constraints_=[single_flip_contiguous, pop_constraint, compactness_bound]
chain = MarkovChain(
proposal=proposal,
constraints=constraints_,
accept=always_accept,
initial_state=initial_partition,
total_steps=num_steps
)
partitions=[] # recording partitions at each step
for index, part in enumerate(chain):
if index % interval == 0:
print('Markov chain step '+str(index))
partitions += [part]
if save_part:
sd.dump_run(settings['partitions_path'], partitions)
print('saved partitions to '+ settings['partitions_path'])
return(partitions)
["D_GOV", "R_GOV"],
["D_ATTGEN", "R_ATTGEN"],
["PRES12D", "PRES12R"],
]
updater = {
"population": updaters.Tally("TAPERSONS", alias="population"),
"cut_edges": cut_edges,
'b_nodes': b_nodes,
}
elections = [
Election(
election_names[i],
{
"First": election_columns[i][0],
"Second": election_columns[i][1]
},
) for i in range(num_elections)
]
election_updaters = {election.name: election for election in elections}
updater.update(election_updaters)
initial_partition = Partition(graph, "HB5005", updater)
print("built initial partition")
ideal_population = sum(
initial_partition["population"].values()) / len(initial_partition)
# print(ideal_population)
edge_color=colors[count])
plt.savefig("va_precinct_" + str(count + 1) + ".png",
bbox_inches='tight',
dpi=400)
plt.clf()
count += 1
plt.savefig("VA_precinct_graph.png", bbox_inches='tight', dpi=400)
#CD.plot(column='CD108FP', cmap='seismic')
real_life_plan = Partition(graph, "CD")
plt.axis('off')
plt.show()
plt.savefig("va_precinct_graph.png", bbox_inches='tight', dpi=400)
#Gerrychain on 2016 election for virginia.
graph.nodes[0]
election = Election("PRES16", {"Dem": "G16DPRS", "Rep": "G16RPRS"})
initial_partition = Partition(graph,
assignment="CD_16",
updaters={
"cut_edges": cut_edges,
"population": Tally("TOTPOP",
alias="population"),
"PRES16": election
})
for district, pop in initial_partition["population"].items():
print("District {}: {}".format(district, pop))
from gerrychain import MarkovChain
from gerrychain.constraints import single_flip_contiguous
from gerrychain.proposals import propose_random_flip
from gerrychain.accept import always_accept
["T16PRESD", "T16PRESR"],
["W101216D", "W101216R"],
]
graph = Graph.from_json(graph_path)
updaters = {
"population": updaters.Tally("TOT_POP", alias="population"),
"cut_edges": cut_edges,
}
elections = [
Election(
election_names[i],
{
"Democratic": election_columns[i][0],
"Republican": election_columns[i][1]
},
) for i in range(num_elections)
]
election_updaters = {election.name: election for election in elections}
updaters.update(election_updaters)
initial_partition = Partition(graph, "2011_PLA_1", updaters)
#Other possible starting plans are:
#"2011_PLA_1": Congressional district ID in 2011 enacted congressional map
#"GOV": Congressional district ID in Governor’s counter-proposed plan
#"TS": Congressional district ID in Turzai-Scarnati Plan
POP_COL = "TOTPOP"
NUM_DISTRICTS = num_districts_in_map[args.map]
ITERS = args.n
EPS = epsilons[args.map]
ELECTS = ["PRES16", "SEN16", "GOV16", "GOV18", "AG16", "SOS16", "USH18"]
## Pull in graph and set up updaters
print("Reading in Data/Graph")
df = gpd.read_file("data/OR_precincts/OR_precincts.shp")
with open("data/OR_precincts/OR_precinct_graph.p", "rb") as f_in:
graph = pickle.load(f_in)
elections = [Election("USH18",{"Dem": "USH18D","Rep":"USH18R"}),]
Election("SEN16", {"Dem": "SEN16D", "Rep": "SEN16R"}),
Election("GOV16", {"Dem": "GOV16D", "Rep": "GOV16R"}),
Election("AG16", {"Dem": "AG16D", "Rep": "AG16R"}),
Election("SOS16", {"Dem": "SOS16D", "Rep": "SOS16R"}),
Election("GOV18", {"Dem": "GOV18D", "Rep": "GOV18R"})]
or_updaters = {"population" : Tally(POP_COL, alias="population"),
"cut_edges": cut_edges}
election_updaters = {election.name: election for election in elections}
or_updaters.update(election_updaters)
## Create seed plans and Set up Markov chain
# load actual gerrychain results
sen_05 = np.load(
"/Users/hopecj/projects/gerryspam/MO/res_0817/MO_state_senate_300000_0.05.p"
)
sen_01 = np.load(
"/Users/hopecj/projects/gerryspam/MO/res_0817/MO_state_senate_100000_0.01.p"
)
sen_05.keys() # shows "columns" available
print(sen_05["mean_median_ussen16"]) # results from chain
# open enacted map
graph = Graph.from_file(
"/Users/hopecj/projects/gerryspam/MO/dat/final_prec/prec_labeled.shp")
elections = [
Election("USSEN16", {
"Dem": "G16USSDKAN",
"Rep": "G16USSRBLU"
}),
Election("PRES16", {
"Dem": "G16PREDCLI",
"Rep": "G16PRERTRU"
})
]
mo_updaters = {
"population": Tally("POP10", alias="population"),
"cut_edges": cut_edges
}
election_updaters = {election.name: election for election in elections}
mo_updaters.update(election_updaters)
sen_part = Partition(graph, assignment="SLDUST", updaters=mo_updaters)
"SENW101216",
]
election_columns = [
["PRES12D", "PRES12R"],
["T16PRESD", "T16PRESR"],
["W101216D", "W101216R"],
]
graph = Graph.from_json(graph_path)
updaters = {
"population": updaters.Tally("TOT_POP", alias="population"),
"cut_edges": cut_edges,
"BPOP":Election("BPOP",{"BPOP":"BPOP","nBPOP":"nBPOP"})
}
elections = [
Election(
election_names[i],
{"Democratic": election_columns[i][0], "Republican": election_columns[i][1]},
)
for i in range(num_elections)
]
election_updaters = {election.name: election for election in elections}
updaters.update(election_updaters)
["D_LTGOV", "R_LTGOV"],
["D_GOV", "R_GOV"],
["D_ATTGEN", "R_ATTGEN"],
["PRES12D", "PRES12R"],
]
updater = {
"population": updaters.Tally("TAPERSONS", alias="population"),
"cut_edges": cut_edges,
'b_nodes':b_nodes,
}
elections = [
Election(
election_names[i],
{"First": election_columns[i][0], "Second": election_columns[i][1]},
)
for i in range(num_elections)
]
election_updaters = {election.name: election for election in elections}
updater.update(election_updaters)
#with open("./Outputs/VA_Old_Tree_Plans/VA_Tree_small0198.json", 'r') as f:
# tree_dict = json.load(f)
#tree_dict = dict(tree_dict)
#for n in graph.nodes():
"cut_edges": cut_edges,
"BVAP": updaters.Tally(BVAP_col, alias = "BVAP"),
"HVAP": updaters.Tally(HVAP_col, alias = "HVAP"),
"AVAP": updaters.Tally(AVAP_col, alias = "AVAP"),
"NVAP": updaters.Tally(Native_col, alias = "NVAP"),
"PVAP": updaters.Tally(Pacific_col, alias = "PVAP"),
"VAP": updaters.Tally(VAP_col, alias = "VAP"),
"county_splits": county_splits("county_splits", county_assignment_col),
"racial_demographics": lambda p: racial_demographics(p, "VAP", "BVAP", "HVAP", "AVAP", "NVAP", "PVAP")
}
num_elections = len(election_names)
elections = [
Election(
election_names[i],
{"Dem": election_columns[i][0], "Rep": election_columns[i][1]},
)
for i in range(num_elections)
]
election_updaters = {election.name: election for election in elections}
updaters.update(election_updaters)
# ---- Past Districting Plan -----
districting_partition = GeographicPartition(graph, district_assignment_col, updaters)
compactness_ref = len(districting_partition["cut_edges"])
county_splits_ref = calc_splits(districting_partition["county_splits"])
racial_vap_ref = districting_partition["racial_demographics"]
