"""
Define defaults for use throughout FPsim
"""
# Imports
import numpy as np
import sciris as sc
import starsim as ss
import fpsim
import fpsim.arrays as fpa
#%% Global defaults
useSI = True
mpy = 12 # Months per year, to avoid magic numbers
eps = 1e-9 # To avoid divide-by-zero
min_age = 15 # Minimum age to be considered eligible to use contraceptive methods
max_age = 99 # Maximum age (inclusive)
max_age_preg = 50 # Maximum age to become pregnant
max_parity = 20 # Maximum number of children to track - also applies to abortions, miscarriages, stillbirths
max_parity_spline = 20 # Used for parity splines
location_registry = {} # Registry for external custom locations
valid_country_locs = ['senegal', 'kenya', 'ethiopia', 'cotedivoire', 'niger', 'nigeria_kano', 'nigeria_kaduna', 'nigeria_lagos', 'pakistan_sindh']
valid_region_locs = {
'ethiopia': ['addis_ababa', 'afar', 'amhara', 'benishangul_gumuz', 'dire_dawa', 'gambela', 'harari', 'oromia', 'snnpr', 'somali', 'tigray']
}
# Parse locations
[docs]
def get_location(location, printmsg=False):
if not location:
print("No location specified. Available locations are: ")
print(", ".join(valid_country_locs))
print("To use model defaults, set test=True.")
print(("To use a custom location, you can construct the sim by passing in a dataloader with a path to where you are keeping your data.\n"
"Example:\n"
" import fpsim as fp\n"
" # Load your own data\n"
" my_data = fp.DataLoader(data_path='path-to-my-data')"
" sim = fp.Sim(dataloader=my_data)"))
raise ValueError('Location must be specified. To use model defaults, set test=True.')
location = location.lower() # Ensure it's lowercase
# External locations override internal ones
if location in location_registry:
return location
# Define valid locations
if location not in valid_country_locs and not any(location in v for v in valid_region_locs.values()):
errormsg = f'Location "{location}" is not currently supported'
raise NotImplementedError(errormsg)
return location
[docs]
def get_dataloader(location, printwarn=True):
""" Return the data loader module """
from . import locations as fplocs
location = get_location(location)
if hasattr(fplocs, location):
dataloader = getattr(fplocs, location).dataloader()
else:
raise NotImplementedError(f'Could not find dataloader for {location}')
return dataloader
[docs]
def get_calib_pars(location, verbose=1):
""" Return the calibration parameters """
from . import locations as fplocs
if location is None:
return
location = get_location(location)
if hasattr(fplocs, location):
calib_pars = getattr(fplocs, location).make_calib_pars()
else:
sc.printv(f'No calibration parameters found for {location}', thisverbose=0, verbose=verbose)
return None
return calib_pars
[docs]
def get_test_defaults():
""" Return the test defaults """
defaults = {
'n_agents': 500,
'start': 2000,
'stop': 2005,
'location': 'senegal',
}
return defaults
# Defaults states and values of any new(born) agent unless initialized with data or other strategy
# or updated during the course of a simulation.
fpmod_states = [
# Contraception
ss.BoolState('on_contra', default=False), # whether she's on contraception
ss.IntArr('method', default=0), # Which method to use. 0 used for those on no method
ss.FloatArr('ti_contra', default=0), # time point at which to set method
ss.FloatArr('barrier', default=0),
ss.BoolState('ever_used_contra', default=False), # Ever been on contraception. 0 for never having used
ss.FloatArr('rel_sus', default=0), # Relative susceptibility to pregnancy, set to 1 for active fecund women
# Sexual and reproductive states, all False by default and set during simulation
ss.BoolState('lam'),
ss.BoolState('pregnant'),
ss.BoolState('fertile'),
ss.BoolState('sexually_active'),
ss.BoolState('sexual_debut'),
ss.BoolState('lactating'),
ss.BoolState('postpartum'),
# Ages of key events
ss.FloatArr('sexual_debut_age', default=-1),
ss.FloatArr('fated_debut', default=-1),
ss.FloatArr('first_birth_age', default=-1),
# Counts of events
ss.FloatArr('parity', default=0), # Number of births including stillbirths
ss.FloatArr('n_births', default=0), # Number of live births
ss.FloatArr('n_stillbirths', default=0), # Number of stillbirths
ss.FloatArr('n_miscarriages', default=0), # Number of miscarriages
ss.FloatArr('n_abortions', default=0), # Number of abortions
ss.FloatArr('n_pregnancies', default=0), # Number of pregnancies, including miscarriages, stillbirths, abortions
ss.FloatArr('months_inactive', default=0), # TODO, what does this store?
ss.FloatArr('short_interval', default=0), # TODO, what does this store?
# Durations and counters
ss.FloatArr('gestation', default=0), # TODO, remove?
ss.FloatArr('remainder_months', default=0), # TODO, remove?
ss.FloatArr('dur_pregnancy', default=0),
ss.FloatArr('dur_postpartum', default=0),
ss.FloatArr('dur_breastfeed', default=0),
ss.FloatArr('dur_breastfeed_total', default=0),
# Timesteps of significant events
ss.FloatArr('ti_conceived'),
ss.FloatArr('ti_pregnant'),
ss.FloatArr('ti_delivery'),
ss.FloatArr('ti_last_delivery'),
ss.FloatArr('ti_live_birth'),
ss.FloatArr('ti_stillbirth'),
ss.FloatArr('ti_postpartum'),
ss.FloatArr('ti_miscarriage'),
ss.FloatArr('ti_abortion'),
ss.FloatArr('ti_stop_postpartum'),
ss.FloatArr('ti_stop_breastfeeding'),
ss.FloatArr('ti_debut'),
ss.FloatArr('ti_dead'),
# Fecundity
ss.FloatArr('personal_fecundity', default=0),
# Add these states to the people object. They are not tracked by timestep in the way other states are, so they
# need to be added manually. Eventually these will become part of a separate module tracking pregnancies and
# pregnancy outcomes.
fpa.TwoDimensionalArr('birth_ages', ncols=max_parity), # Ages at time of live births
fpa.TwoDimensionalArr('stillborn_ages', ncols=max_parity), # Ages at time of stillbirths
fpa.TwoDimensionalArr('miscarriage_ages', ncols=max_parity), # Ages at time of miscarriages
fpa.TwoDimensionalArr('abortion_ages', ncols=max_parity), # Ages at time of abortions
ss.BoolState('partnered', default=False), # Will remain at these values if use_partnership is False
ss.FloatArr('partnership_age', default=-1), # Will remain at these values if use_partnership is False
# ss.State('urban', default=True), # Urban/rural
# ss.FloatArr('wealthquintile', default=3), # Wealth quintile
]
# Postpartum keys to months
postpartum_map = {
'pp0to5': [0, 6],
'pp6to11': [6, 12],
'pp12to23': [12, 24]
}
# Age bins for tracking age-specific fertility rate
age_bin_map = {
'10-14': [10, 15],
'15-19': [15, 20],
'20-24': [20, 25],
'25-29': [25, 30],
'30-34': [30, 35],
'35-39': [35, 40],
'40-44': [40, 45],
'45-49': [45, 50]
}
# Age and parity splines
spline_ages = np.arange(max_age + 1)
spline_preg_ages = np.arange(max_age_preg + 1)
spline_parities = np.arange(max_parity_spline + 1)
# Define allowable keys to select all (all ages, all methods, etc)
none_all_keys = [None, 'all', ':', [None], ['all'], [':']]
# Age bins for different method switching matrices -- can be overwritten by locations
method_age_map = {
'<18': [ 0, 18],
'18-20': [18, 20],
'20-25': [20, 25],
'25-35': [25, 35],
'>35': [35, max_age+1], # +1 since we're using < rather than <=
}
immutable_method_age_map = {
'<18': [ 0, 18],
'18-20': [18, 20],
'20-25': [20, 25],
'25-30': [25, 30],
'30-35': [30, 35],
'>35': [35, max_age+1], # +1 since we're using < rather than <=
}
method_youth_age_map = {
'<16': [10, 16],
'16-17': [16, 18],
'18-19': [18, 20],
'20-22': [20, 23],
'23-25': [23, 26],
'>25': [26, max_age+1]
}
# Counts - we compute number of new events each timestep, plus number of cumulative events
event_counts = sc.autolist(
'births',
'stillbirths',
'miscarriages',
'abortions',
'short_intervals',
'secondary_births',
'pregnancies',
'total_births',
'maternal_deaths',
'infant_deaths',
'method_failures',
)
people_counts = sc.autolist(
'contra_access',
'new_users',
'ever_used_contra',
'switchers',
'n_fecund',
'pp0to5',
'pp6to11',
'pp12to23',
'parity0to1',
'parity2to3',
'parity4to5',
'parity6plus',
'nonpostpartum',
)
sim_results = sc.autolist(
'n_urban',
'n_wq1',
'n_wq2',
'n_wq3',
'n_wq4',
'n_wq5',
)
# Rates and other results that aren't scaled
rate_results = sc.autolist(
'tfr',
'mmr',
'imr',
'p_short_interval',
)
