def test_generate_larger_households(location='seattle_metro',
state_location='Washington',
country_location='usa'):
Nhomes_to_sample_smooth = 1000
household_size_distr = sp.get_household_size_distr(datadir, location,
state_location,
country_location)
hh_sizes = sp.generate_household_sizes(Nhomes_to_sample_smooth,
household_size_distr)
hha_brackets = sp.get_head_age_brackets(datadir,
country_location=country_location)
hha_by_size_counts = sp.get_head_age_by_size_distr(
datadir, country_location=country_location)
age_brackets_filepath = sp.get_census_age_brackets_path(
datadir, state_location, country_location)
age_brackets = sp.get_age_brackets_from_df(age_brackets_filepath)
age_by_brackets_dic = sp.get_age_by_brackets_dic(age_brackets)
contact_matrix_dic = sp.get_contact_matrix_dic(
datadir, sheet_name='United States of America')
single_year_age_distr = {}
for n in range(101):
single_year_age_distr[n] = float(1.0 / 101.0)
# generate households of size 3
size = 3
# first variable is the household size to be created, so here this means we want to create all households of size 3 and the hh_sizes variable tells us how many of size 3 will be created at index 3-1 (since hh_sizes is an array rather than a dictionary)
larger_households = sp.generate_larger_households(
size, hh_sizes, hha_by_size_counts, hha_brackets, age_brackets,
age_by_brackets_dic, contact_matrix_dic, single_year_age_distr)
assert larger_households is not None
print(larger_households)
def test_generate_all_households(location='seattle_metro',
state_location='Washington',
country_location='usa'):
N = 1000
household_size_distr = sp.get_household_size_distr(datadir, location,
state_location,
country_location)
hh_sizes = sp.generate_household_sizes_from_fixed_pop_size(
N, household_size_distr)
hha_brackets = sp.get_head_age_brackets(datadir,
country_location=country_location)
hha_by_size_counts = sp.get_head_age_by_size_distr(
datadir, country_location=country_location)
age_brackets_filepath = sp.get_census_age_brackets_path(
datadir, state_location, country_location)
age_brackets = sp.get_age_brackets_from_df(age_brackets_filepath)
age_by_brackets_dic = sp.get_age_by_brackets_dic(age_brackets)
contact_matrix_dic = sp.get_contact_matrix_dic(
datadir, sheet_name='United States of America')
single_year_age_distr = {}
for n in range(101):
single_year_age_distr[n] = float(1.0 / 101.0)
homes_dic, homes = sp.generate_all_households(
N, hh_sizes, hha_by_size_counts, hha_brackets, age_brackets,
age_by_brackets_dic, contact_matrix_dic, single_year_age_distr)
assert homes_dic, homes is not None
syn_ages, syn_sexes = sp.get_usa_age_sex_n(location, state_location,
totalpop)
syn_age_count = Counter(syn_ages)
syn_age_distr = sp.norm_dic(Counter(syn_ages))
N = Nhomes
hh_sizes = sp.generate_household_sizes_from_fixed_pop_size(
N, household_size_distr)
totalpop = sp.get_totalpopsize_from_household_sizes(hh_sizes)
print(totalpop, 'pop')
hha_df = sp.get_household_head_age_by_size_df(datadir, country_location,
use_bayesian)
hha_brackets = sp.get_head_age_brackets(datadir, country_location,
use_bayesian)
hha_by_size = sp.get_head_age_by_size_distr(datadir, country_location,
use_bayesian)
homes_dic, homes = sp.generate_all_households(hh_sizes, hha_by_size,
hha_brackets, age_brackets,
age_by_brackets_dic,
contact_matrix_dic,
syn_age_distr)
c = 0
for s in homes_dic:
c += s * len(homes_dic[s])
print('c', c)
homes_by_uids, age_by_uid_dic = sp.assign_uids_by_homes(homes)
def check_household_head(pop,
n,
datadir,
figdir,
state_location=None,
country_location=None,
file_path=None,
use_default=False,
test_prefix="",
do_close=True):
"""
Check the household head by age distribution against the reference data
Args:
pop : population dictionary
n : population size
datadir : root data directory which has resides the reference data
figdir : directory where to result files are saved
state_location : name of the state the location is in
country_location : name of the country the location is in
file_path : file path to user specified gender by age bracket distribution data
use_default : if True, try to first use the other parameters to find data specific to the location
under study, otherwise returns default data drawing from Seattle, Washington.
test_prefix : used for prefix of the plot title
do_close : close the image immediately if set to True
Returns:
None.
Plots will be save to figdir if provided
"""
figdir = os.path.join(figdir, "household_head")
household_head_age_distribution_by_family_size = sp.get_head_age_by_size_distr(
state_location=state_location, country_location=country_location)
head_age_brackets = sp.get_head_age_brackets(
state_location=state_location, country_location=country_location)
# Inverse the mapping for use below
hha_index = sp.get_index_by_brackets_dic(head_age_brackets)
household_head_age_distribution_by_family_size = household_head_age_distribution_by_family_size[
1:]
expected_hh_ages = pandas.DataFrame(
household_head_age_distribution_by_family_size)
expected_hh_ages_percentage = expected_hh_ages.div(
expected_hh_ages.sum(axis=0), axis=1)
actual_hh_ages_percetnage = utilities.get_household_head_age_size(
pop, index=hha_index)
expected_values = expected_hh_ages_percentage.values[1:, :]
actual_values = actual_hh_ages_percetnage.values
xlabels = [
f'{min(head_age_brackets[bracket_index])}-{max(head_age_brackets[bracket_index])}'
for bracket_index in head_age_brackets.keys()
]
family_sizes = [i + 2 for i in range(0, len(expected_hh_ages_percentage))]
utilities.plot_heatmap(
expected_values,
actual_values,
xlabels,
family_sizes,
'Head of Household Age',
'Household Size',
# expected_hh_ages_percentage.columns, # family_sizes,
testprefix="household_head_age_family_size " + test_prefix,
figdir=figdir,
do_close=do_close)
syn_ages, syn_sexes = sp.get_usa_age_sex_n(datadir, location,
state_location,
country_location, totalpop)
syn_age_count = Counter(syn_ages)
syn_age_distr = sp.norm_dic(Counter(syn_ages))
N = Nhomes
hh_sizes = sp.generate_household_sizes_from_fixed_pop_size(
N, household_size_distr)
totalpop = sp.get_totalpopsize_from_household_sizes(hh_sizes)
print(totalpop, 'pop')
hha_df = sp.get_household_head_age_by_size_df(datadir, state_location,
country_location)
hha_brackets = sp.get_head_age_brackets(datadir,
country_location=country_location)
hha_by_size = sp.get_head_age_by_size_distr(
datadir, country_location=country_location)
homes_dic, homes = sp.generate_all_households(N, hh_sizes, hha_by_size,
hha_brackets, age_brackets,
age_by_brackets_dic,
contact_matrix_dic,
deepcopy(syn_age_distr))
homes_by_uids, age_by_uid_dic = sp.assign_uids_by_homes(homes)
new_ages_count = Counter(age_by_uid_dic.values())
fig = plt.figure(figsize=(6, 4))
ax = fig.add_subplot(111)
x = np.arange(100)