def single_species_sex():
with pd.Domain('seven_kingdoms.popdyn', csx=1., csy=1., shape=shape, top=shape[0], left=0) as domain:
domain.add_carrying_capacity(starks, stark_k, 0, stark_k_data, distribute=False)
domain.add_population(starks, 10000., 0, distribute_by_habitat=True)
domain.add_population(pd.Sex('Stark', 'male'), 5000., 0, distribute_by_habitat=True)
domain.add_population(pd.Sex('Stark', 'female'), 5000., 0, distribute_by_habitat=True)
pd.solvers.discrete_explicit(domain, 0, 2).execute()
# Population should remain consistent at 10,000 for time steps 1 and 2
for i in range(1, 3):
tot_pop = summary.total_population(domain, 'stark', i).sum()
if not np.isclose(tot_pop, 20000):
raise Exception('The total population has changed to {} at time {}'.format(tot_pop, i))
tot_pop = summary.total_population(domain, 'stark', i, 'male').sum()
if not np.isclose(tot_pop, 10000):
raise Exception('The male population has changed to {} at time {}'.format(tot_pop, i))
tot_pop = summary.total_population(domain, 'stark', i, 'female').sum()
if not np.isclose(tot_pop, 10000):
raise Exception('The female population has changed to {} at time {}'.format(tot_pop, i))
cc = summary.total_carrying_capacity(domain, 'stark', i, 'male').sum()
if not np.isclose(cc, (stark_k_data).sum()):
raise Exception('The male carrying capacity has changed to {} at time {}'.format(cc, i))
cc = summary.total_carrying_capacity(domain, 'stark', i, 'male').sum()
if not np.isclose(cc, (stark_k_data).sum()):
raise Exception('The female carrying capacity has changed to {} at time {}'.format(cc, i))
def conversion_and_transmission():
# Pasted from recipient species
with pd.Domain('seven_kingdoms.popdyn', csx=1., csy=1., shape=(1, 1), top=shape[0], left=0) as domain:
recipient = pd.Species('recipient')
test_mortality = pd.Mortality('test recipient')
test_mortality.add_recipient_species(recipient)
test_file = os.path.join(os.path.dirname(__file__), 'direct_transmission.csv')
starks.add_disease(test_file, **{
'direct_transmission': True,
'environmental_transmission': True,
'E_data': {
1: np.array([[0.1]]),
2: np.array([[0.1]])
}
})
domain.add_carrying_capacity(starks, stark_k, 0, 10000, distribute=False)
domain.add_population(starks, 10000., 0, distribute_by_habitat=True)
domain.add_mortality(starks, test_mortality, 0, 0.1)
domain.add_population(recipient, 0, 10)
pd.solvers.discrete_explicit(domain, 0, 2).execute()
Pi = 10 / 10000
FOI = 0.1 * Pi
env = 0.1 * 0.1
conv = 10000 * (FOI + env)
tot_pop = summary.total_population(domain, 'recipient', 1).sum()
if not np.isclose(tot_pop, conv):
raise Exception('Recipient population should be {}, got {} at time 1'.format(
conv, tot_pop)
)
def rate_based_mortality():
# Pasted from recipient species
with pd.Domain('seven_kingdoms.popdyn', csx=1., csy=1., shape=(1, 1), top=shape[0], left=0) as domain:
recipient = pd.Species('recipient')
time_based_mortality = pd.Mortality('time based')
time_based_mortality.add_time_based_mortality([0.2, 0.5, 1.])
domain.add_mortality(recipient, time_based_mortality, 0)
test_mortality = pd.Mortality('test recipient')
test_mortality.add_recipient_species(recipient)
domain.add_carrying_capacity(starks, stark_k, 0, 10000, distribute=False)
domain.add_population(starks, 10000., 0, distribute_by_habitat=True)
domain.add_mortality(starks, test_mortality, 0, 0.1)
pd.solvers.discrete_explicit(domain, 0, 5).execute()
tot_shouldbe = [1000, 900, 810, 729, 656.10004]
mort_shouldbe = [0, 200, 450, 810, 0]
tot_pop, mort = [], []
for year in range(1, 6):
tot_pop.append(summary.total_population(domain, 'recipient', year).sum())
mort.append(np.squeeze(summary.total_mortality(domain, 'recipient', year, mortality_name='time based')))
if not np.allclose(tot_pop, tot_shouldbe) or not np.allclose(mort, mort_shouldbe):
raise Exception('Population should be {}, got {}\nMortality should be {}, got {}'.format(
tot_shouldbe, tot_pop, mort_shouldbe, mort)
)
def species_as_mortality():
# White Walker mortality lookup: [(0, 0), (0.1, 0.1), (0.5, 0.8), (1, 0.9)]
with pd.Domain('seven_kingdoms.popdyn', csx=1., csy=1., shape=(1, 1), top=shape[0], left=0) as domain:
domain.add_carrying_capacity(starks, stark_k, 0, 100)
domain.add_population(starks, 100, 0)
domain.add_mortality(starks, white_walker_death, 0)
# 30% density, mortality should be 0.45
domain.add_carrying_capacity(white_walker, white_walker_k, 0, 100)
domain.add_population(white_walker, 30., 0)
# 50% density, mortality should be 0.8
domain.add_carrying_capacity(white_walker, white_walker_k, 2, 60)
pd.solvers.discrete_explicit(domain, 0, 2).execute()
tot_pop = summary.total_population(domain, 'stark', 1).sum()
if not np.isclose(tot_pop, 100 - (100 * 0.45)):
raise Exception('Population is {}, but should be {} at time step 1'.format(tot_pop, 100 - (100 * 0.45)))
tot_pop = summary.total_population(domain, 'stark', 2).sum()
if not np.isclose(tot_pop, 55. - (55 * 0.8)):
raise Exception('Population is {}, but should be {} at time step 1'.format(tot_pop, 55. - (55 * 0.8)))
def single_species_recipient():
with pd.Domain('seven_kingdoms.popdyn', csx=1., csy=1., shape=(1, 1), top=shape[0], left=0) as domain:
recipient = pd.Species('recipient')
test_mortality = pd.Mortality('test recipient')
test_mortality.add_recipient_species(recipient)
domain.add_carrying_capacity(starks, stark_k, 0, 10000, distribute=False)
domain.add_population(starks, 10000., 0, distribute_by_habitat=True)
domain.add_mortality(starks, test_mortality, 0, 0.1)
domain.add_population(recipient, 0, 10)
pd.solvers.discrete_explicit(domain, 0, 2).execute()
tot_pop = summary.total_population(domain, 'recipient', 1).sum()
if tot_pop != 10000 * .1:
raise Exception('Recipient population should be {}, got {} at time 1'.format(
10000 * .1, tot_pop)
)
tot_pop = summary.total_population(domain, 'recipient', 2).sum()
if tot_pop != (10000 - (10000 * .1)) * .1:
raise Exception('Recipient population should be {}, got {} at time 2'.format(
(10000 - (10000 * .1)) * .1, tot_pop)
)
def single_species_emigration():
with pd.Domain('seven_kingdoms.popdyn', csx=1., csy=1., shape=shape, top=shape[0], left=0) as domain:
domain.add_carrying_capacity(starks, stark_k, 0, stark_k_data, distribute=False)
domain.add_population(starks, 10000., 0, distribute_by_habitat=True)
# Add population at time 1
domain.add_population(starks, 10000., 1, distribute_by_habitat=True)
pd.solvers.discrete_explicit(domain, 0, 2).execute()
# Population should 20,000 at time step 1
tot_pop = summary.total_population(domain, 'stark', 1).sum()
if not np.isclose(tot_pop, 20000):
raise Exception('The population should be {}, got {}'.format(20000, tot_pop))
def single_species_agegroups():
with pd.Domain('seven_kingdoms.popdyn', csx=1., csy=1., shape=shape, top=shape[0], left=0) as domain:
pops = [250., 250., 600., 600., 3000., 3000.]
tot_pop = np.sum(pops)
domain.add_carrying_capacity(starks, stark_k, 0, stark_k_data + tot_pop, distribute=False)
domain.add_population(stark_male_infant, 250., 0, distribute_by_habitat=True)
domain.add_population(stark_female_infant, 250., 0, distribute_by_habitat=True)
domain.add_population(stark_male_adolescent, 600., 0, distribute_by_habitat=True)
domain.add_population(stark_female_adolescent, 600., 0, distribute_by_habitat=True)
domain.add_population(stark_male_adult, 3000., 0, distribute_by_habitat=True)
domain.add_population(stark_female_adult, 3000., 0, distribute_by_habitat=True)
pd.solvers.discrete_explicit(domain, 0, 2).execute()
for i in range(1, 3):
_pop = summary.total_population(domain, 'stark', i).sum()
# Subtract old age mortality for the total population
if not np.isclose(_pop, tot_pop - 3000. / (50 - 13. + 1) * 2 * i):
raise Exception('The population should be {}, not {} at time {}'.format(
tot_pop - 3000. / (50 - 13. + 1) * 2, _pop, i)
)
# Check propagation
_pop = summary.total_population(domain, 'stark', i, 'male', group=stark_male_infant.group_name).sum()
if not np.isclose(_pop, 250 - ((250 / 4.) * i)):
raise Exception('{} population should be {}, but is {} at time {}'.format(
stark_male_infant.group_name, 250 - ((250 / 4.) * i), _pop, i
))
_pop = summary.total_population(domain, 'stark', i, 'male', group=stark_male_adolescent.group_name).sum()
if not np.isclose(_pop, 600 - ((600 / 9.) * i) + ((250 / 4.) * i)):
raise Exception('{} population should be {}, but is {} at time {}'.format(
stark_male_adolescent.group_name, 600 - ((600 / 9.) * i) + ((250 / 4.) * i), _pop, i
))
_pop = summary.total_population(domain, 'stark', i, 'male', group=stark_male_adult.group_name).sum()
if not np.isclose(_pop, 3000. - (3000. / (50 - 13. + 1) * i) + ((600 / 9.) * i)):
raise Exception('{} population should be {}, but is {} at time {}'.format(
stark_male_adult.group_name, 3000. - (3000. / (50 - 13. + 1) * i) + ((600 / 9.) * i), _pop, i
))
def global_n_interspecies():
# White Walker mortality lookup: [(0, 0), (0.1, 0.1), (0.5, 0.8), (1, 0.9)]
with pd.Domain('seven_kingdoms.popdyn', csx=1., csy=1., shape=(1, 1), top=shape[0], left=0) as domain:
domain.add_carrying_capacity(starks, stark_k, 0, 100)
domain.add_population(starks, 100, 0)
white_walker_death = pd.Mortality('White Walker')
white_walker_death.add_as_species(white_walker, [(0, 0), (1/3., 0.1), (0.35714, 0.5), (1, 1)],
population_type='global ratio')
domain.add_mortality(starks, white_walker_death, 0)
# should be 1/3, making mortality 0.1
domain.add_carrying_capacity(white_walker, white_walker_k, 0, 100)
domain.add_population(white_walker, 50., 0)
pd.solvers.discrete_explicit(domain, 0, 2, global_density=True).execute()
tot_pop = summary.total_population(domain, 'stark', 1).sum()
if not np.isclose(tot_pop, 90):
raise Exception('Population is {}, but should be {} at time step 1'.format(tot_pop, 90))
tot_pop = summary.total_population(domain, 'stark', 2).sum()
if not np.isclose(tot_pop, 90 * .5):
raise Exception('Population is {}, but should be {} at time step 2'.format(tot_pop, 90 * .5))
def single_species_mask():
with pd.Domain('seven_kingdoms.popdyn', csx=1., csy=1., shape=shape, top=shape[0], left=0) as domain:
starks = pd.Species('Stark')
stark_male_adolescent = pd.AgeGroup('Stark', 'Adolescent', 'male', 4, 12)
stark_male_adult = pd.AgeGroup('Stark', 'Adult', 'male', 13, 50)
stark_male_adolescent.add_dispersal('masked density-based dispersion', (10,))
domain.add_carrying_capacity(starks, stark_k, 0, stark_k_data, distribute=False)
domain.add_population(stark_male_adolescent, np.random.random(shape), 0, distribute=False)
domain.add_population(stark_male_adult, np.random.random(shape), 0, distribute=False)
domain.add_mask(starks, 0, np.random.random(shape), distribute=False) # Should be inherited
pd.solvers.discrete_explicit(domain, 0, 2).execute()
for i in range(1, 2):
_pop = summary.total_population(domain, 'stark', i).sum() + summary.total_mortality(domain, 'stark', i).sum()
if not np.isclose(_pop, summary.total_population(domain, 'stark', 0).sum()):
raise Exception('Population should be {}, got {}'.format(
summary.total_population(domain, 'stark', 0).sum(), _pop)
)
if np.allclose(summary.total_population(domain, 'stark', i) + summary.total_mortality(domain, 'stark', i),
summary.total_population(domain, 'stark', 0)):
raise Exception('No dispersal occurred')
def single_species_mvp():
with pd.Domain('seven_kingdoms.popdyn', csx=1., csy=1., shape=shape, top=shape[0], left=0) as domain:
stark_with_mvp = pd.Species('stark', minimum_viable_population=100, minimum_viable_area=shape[0] * shape[1])
domain.add_carrying_capacity(stark_with_mvp, stark_k, 0, shape[0] * shape[1] * 100.)
domain.add_population(stark_with_mvp, 50., 0)
pd.solvers.discrete_explicit(domain, 0, 10).execute()
# Population should be reduced
tp = []
for i in range(0, 11):
tp.append(summary.total_population(domain, 'stark', i).sum())
if not any([t == 0 for t in tp]):
raise Exception('The population did not succumb to MVP (populations: {})'.format(tp))
def reverse_conversion_at_birth():
with pd.Domain('seven_kingdoms.popdyn', csx=1., csy=1., shape=(1, 1), top=shape[0], left=0) as domain:
recipient_males = pd.Sex('recipient', 'male')
recipient_females = pd.Sex('recipient', 'female')
test_mortality = pd.Mortality('test recipient')
test_mortality.add_recipient_species(recipient_females)
domain.add_population(recipient_males, 10, 0)
domain.add_population(recipient_females, 10, 0)
domain.add_fecundity(recipient_males, pd.Fecundity('test'), 0, .05)
domain.add_fecundity(recipient_females, pd.Fecundity('test'), 0, .05)
domain.add_carrying_capacity(starks, stark_k, 0, 10000, distribute=False)
domain.add_population(pd.Sex('Stark', 'male'), 0.5, 0, distribute_by_habitat=True)
domain.add_population(pd.Sex('Stark', 'female'), 0.5, 0, distribute_by_habitat=True)
domain.add_mortality(starks, test_mortality, 0, 0.1)
pd.solvers.discrete_explicit(domain, 0, 2).execute()
# There should be 2 starks at time 1
tot_pop = summary.total_population(domain, 'stark', 1).sum()
if tot_pop != 2:
raise Exception('No offspring from recipient species in Starks at time 1')
def single_species_dispersion():
for _iter, method in enumerate(['density-based dispersion', 'distance propagation']):
with pd.Domain('seven_kingdoms.popdyn', csx=1., csy=1., shape=shape, top=shape[0], left=0) as domain:
starks = pd.Species('Stark')
starks.add_dispersal(method, (10,))
# Avoid density-dependent mortality
domain.add_carrying_capacity(starks, stark_k, 0, stark_k_data + 1000., distribute=False)
seed_population = np.zeros(shape=shape, dtype='float32')
seed_population[(np.random.randint(0, shape[0], 10), np.random.randint(0, shape[1], 10))] = 100
domain.add_population(starks, seed_population, 0, distribute=False)
try:
pd.solvers.discrete_explicit(domain, 0, 2).execute()
except NotImplementedError:
domain.file.close()
if os.path.isfile('seven_kingdoms.popdyn'):
os.remove('seven_kingdoms.popdyn')
else:
shutil.rmtree('seven_kingdoms.popdyn')
continue
prv_pop = seed_population
for i in range(1, 3):
_pop = summary.total_population(domain, 'stark', i)
if not np.isclose(_pop.sum(), prv_pop.sum()):
raise Exception('The population changed from {} to {} at time {}'.format(prv_pop.sum(), _pop.sum() ,i))
if np.allclose(_pop, prv_pop):
raise Exception('The population did not disperse using {}'.format(method))
prv_pop = _pop
if _iter == 0:
if os.path.isfile('seven_kingdoms.popdyn'):
os.remove('seven_kingdoms.popdyn')
else:
shutil.rmtree('seven_kingdoms.popdyn')
def single_species_random_k():
with pd.Domain('seven_kingdoms.popdyn', csx=1., csy=1., shape=shape, top=shape[0], left=0) as domain:
rand_k = pd.CarryingCapacity('Stark Habitat')
rand_k_data = some_random_k(shape, 1.)
rand_k.random('normal', **{'scale': 10})
domain.add_carrying_capacity(starks, rand_k, 0, rand_k_data, distribute=False)
domain.add_population(starks, rand_k_data, 0, distribute=False)
pd.solvers.discrete_explicit(domain, 0, 2).execute()
# Population is at k, so any k above the previous year should result in:
# 1. density dependent mortality
# 2. a reciprocal reduction in population
prv_pop = rand_k_data
for i in range(1, 3):
# Make sure k is variable
cc = summary.total_carrying_capacity(domain, 'stark', i)
if np.all(cc == rand_k_data):
raise Exception('The k is unchanged')
tot_pop = summary.total_population(domain, 'stark', i)
dd_mort = summary.total_mortality(domain, 'stark', i, mortality_name='Density Dependent')
if not np.allclose(-dd_mort, (tot_pop - prv_pop), atol=1E-04):
raise Exception('DD mortality not correct at time {}'.format(i))
prv_pop = tot_pop
def single_species_mortality():
with pd.Domain('seven_kingdoms.popdyn', csx=1., csy=1., shape=(1, 1), top=shape[0], left=0) as domain:
# Test old age propagation here
stark_male_adolescent = pd.AgeGroup('Stark', 'Adolescent', 'male', 4, 12, live_past_max=True)
stark_female_adolescent = pd.AgeGroup('Stark', 'Adolescent', 'female', 4, 12, live_past_max=True)
domain.add_carrying_capacity(starks, stark_k, 0, 1000., distribute=False)
domain.add_population(stark_male_adolescent, 10., 0)
domain.add_population(stark_female_adolescent, 10., 0)
domain.add_mortality(stark_male_adolescent, disease, 0, 0.2)
domain.add_mortality(stark_female_adolescent, disease, 0, 0.2)
domain.add_mortality(stark_female_adolescent, accident, 0, 0.1)
domain.add_mortality(stark_male_adolescent, accident, 0, 0.1)
pd.solvers.discrete_explicit(domain, 0, 2).execute()
for sex in ['male', 'female']:
prv_pop = 10
for i in range(1, 3):
tot_pop = summary.total_population(domain, 'stark', i, sex, group='adolescent')
mort1 = summary.total_mortality(domain, 'stark', i, sex, mortality_name='Accident')
mort2 = summary.total_mortality(domain, 'stark', i, sex, mortality_name='Disease')
if not np.isclose(tot_pop.sum(), prv_pop - (mort1.sum() + mort2.sum())):
raise Exception('{} adolescent mortality not correct at time {}'.format(sex, i))
prv_pop = tot_pop
def single_species_fecundity():
# F:M ratio allows offspring
with pd.Domain('seven_kingdoms.popdyn', csx=1., csy=1., shape=(1, 1), top=shape[0], left=0) as domain:
domain.add_carrying_capacity(starks, stark_k, 0, 1000., distribute=False)
males = pd.Sex('Stark', 'male')
females = pd.Sex('Stark', 'female')
domain.add_population(males, 50, 0)
domain.add_population(females, 50, 0)
domain.add_fecundity(males, male_fecundity, 0, 1.1)
domain.add_fecundity(females, female_fecundity, 0, 1.1)
pd.solvers.discrete_explicit(domain, 0, 2).execute()
# A constant set of offspring should result
prv_pop = 50.
for i in range(1, 3):
tot_pop = summary.total_population(domain, 'stark', i, 'female').sum()
tot_off = summary.total_offspring(domain, 'stark', i, 'female').sum()
if not np.isclose(tot_off, prv_pop * 1.1):
raise Exception(
'The offspring is {} and should be {} at time {}'.format(tot_off, prv_pop * 1.1, i)
)
if not np.isclose(tot_pop, prv_pop + (prv_pop * 1.1) / 2):
raise Exception(
'The population is {} and should be {} at time {}'.format(
tot_pop, prv_pop + (prv_pop * 1.1) / 2, i
)
)
prv_pop = tot_pop
if os.path.isfile('seven_kingdoms.popdyn'):
os.remove('seven_kingdoms.popdyn')
else:
shutil.rmtree('seven_kingdoms.popdyn')
# F:M ratio disallows offspring
with pd.Domain('seven_kingdoms.popdyn', csx=1., csy=1., shape=(1, 1), top=shape[0], left=0) as domain:
domain.add_carrying_capacity(starks, stark_k, 0, 60)
stark_male = pd.Sex('Stark', 'male')
stark_female = pd.Sex('Stark', 'female')
domain.add_population(stark_male, 20., 0)
domain.add_population(stark_female, 40., 0)
test_female_fecundity_1 = pd.Fecundity('females', fecundity_lookup=[(0., 1.), (2., 0.)])
# Explore density-dependence
test_female_fecundity_2 = pd.Fecundity('females',
density_fecundity_threshold=0.9,
density_fecundity_max=1.1,
fecundity_lookup=[(0., 1.), (1.5, 1. / 1.1)])
domain.add_fecundity(stark_male, male_fecundity, 0, 1.0)
domain.add_fecundity(stark_female, test_female_fecundity_1, 0, 1.1)
domain.add_fecundity(stark_female, test_female_fecundity_2, 2, 1.1)
pd.solvers.discrete_explicit(domain, 0, 2).execute()
male_pop = summary.total_offspring(domain, 'stark', 1, 'male').sum()
if male_pop != 0:
raise Exception('Offspring is {} when should be 0'.format(male_pop))
female_pop = summary.total_population(domain, 'stark', 2, 'female').sum()
# Fecundity should be 0.5
if not np.isclose(female_pop, 40 + (40 * 0.5 / 2)):
raise Exception(
'The population is {} and should be {}'.format(female_pop, 40 + (40 * 0.5 / 2))
)