import CalibrationClasses as Cls
import CalibrationSettings as CalibSets
import SimPy.FigureSupport as Fig
# create a calibration object
calibration = Cls.Calibration()
# sample the posterior of the mortality probability
calibration.sample_posterior(n_samples=CalibSets.POST_N)
# create the histogram of the resampled mortality probabilities
Fig.graph_histogram(data=calibration.mortalityResamples,
title='Histogram of Resampled Mortality Probabilities',
x_label='Mortality Probability',
y_label='Counts',
x_range=[CalibSets.POST_L, CalibSets.POST_U])
# Estimate of mortality probability and the posterior interval
print(
'Estimate of mortality probability ({:.{prec}%} credible interval):'.
format(1 - CalibSets.ALPHA, prec=0),
calibration.get_mortality_estimate_credible_interval(
alpha=CalibSets.ALPHA))
# effective sample size
# txtEff = 'Effective sample size: {:.1f}'.format(calibration.get_effective_sample_size())
# print(txtEff)
import CalibrationClasses as Cls
calibrated = Cls.Calibration()
calibrated.sample_posterior()
print('PROBLEM 4:')
print(
'The estimated annual mortality probability and the 95% credible interval is:',
calibrated.get_mortality_estimate_credible_interval(alpha=.05, deci=4))
print('Problem 2')
print(
'Binomial distribution with p (probability of successes) equals the probability of surviving beyond 5 years (q) and number of trials as N.'
)
print()
print('Problem 3')
prob = stat.binom.pmf(400, 573, 0.5)
print(
'The likelihood that a clinical study reports 400 of 573 participants survived at the end of the 5-year study period '
'if 50% of the patients in our simulated cohort survived beyond 5 years is:',
prob)
print()
print("Problem 4")
# create a calibration object
calibration = CalibClasses.Calibration()
# sample the posterior of the mortality probability
calibration.sample_posterior()
# create the histogram of the resampled mortality probabilities
Fig.graph_histogram(observations=calibration.get_mortality_resamples(),
title='Histogram of Resampled Mortality Probabilities',
x_label='Mortality Probability',
y_label='Counts',
x_range=[CalibSets.POST_L, CalibSets.POST_U])
# Estimate of mortality probability and the posterior interval
print(
'Estimate of mortality probability ({:.{prec}%} credible interval):'.
format(1 - CalibSets.ALPHA, prec=0),
import CalibrationClasses as CalibSupport
import CalibrationSettings as CalibSets
print('(Question 5)')
# initialize a calibrated model
calibrated_model = CalibSupport.CalibratedModel('CalibrationResults.csv')
# simulate the calibrated model
calibrated_model.simulate(CalibSets.SIM_POP_SIZE, CalibSets.TIME_STEPS)
# report mean and projection interval
print(
'Mean survival time and {:.{prec}%} projection interval:'.format(
1 - CalibSets.ALPHA, prec=0),
calibrated_model.get_mean_survival_time_proj_interval(CalibSets.ALPHA,
deci=4))
import CalibrationClasses as Cls
import InputData as D
import SupportCalibrated as Support
from definitions import CALIBRATION_ROOT_DIR
# create a calibrated model for when drug is not available
calibratedModelNoDrug = Cls.CalibratedModel(
csv_file_name=CALIBRATION_ROOT_DIR + '\CalibrationResults.csv')
# simulate the calibrated model
calibratedModelNoDrug.simulate(num_of_simulated_cohorts=D.NUM_SIM_COHORTS,
cohort_size=D.SIM_POP_SIZE,
time_steps=D.TIME_STEPS)
# create a calibrated model when drug is available
calibratedModelWithDrug = Cls.CalibratedModel(
csv_file_name=CALIBRATION_ROOT_DIR + '\CalibrationResults.csv',
drug_effectiveness_ratio=D.DRUG_EFFECT_RATIO)
# simulate the calibrated model
initial_seed = D.NUM_SIM_COHORTS * D.SIM_POP_SIZE
calibratedModelWithDrug.simulate(num_of_simulated_cohorts=D.NUM_SIM_COHORTS,
cohort_size=D.SIM_POP_SIZE,
time_steps=D.TIME_STEPS,
cohort_ids=range(
initial_seed,
initial_seed + D.NUM_SIM_COHORTS))
# report mean and projection interval of expected survival time
Support.print_outcomes(calibrated_model=calibratedModelNoDrug,
strategy_name='When drug is not available:')
Support.print_outcomes(calibrated_model=calibratedModelWithDrug,
strategy_name='When drug is available:')
#P3: Likelihood Calculation
import scipy.stats as stat
from scipy.stats import binom
print(
"Problem3: The likelihood that a clinical study reports 400 of 573 participants survived at the end of the 5-year study period if 50% of the patients in our simulated cohort survived beyond 5 years is",
stat.binom.pmf(k=400, n=573, p=0.5))
#P4: Calibration
import CalibrationClasses as Cls
import CalibrationSettings as CalibSets
# create a calibration object
calibration = Cls.Calibration()
# sample the posterior of the mortality probability
calibration.sample_posterior()
# Estimate of mortality probability and the posterior interval
print(
'Problem4: Estimate of annual mortality probability ({:.{prec}%} credible interval):'
.format(1 - CalibSets.ALPHA, prec=0),
calibration.get_mortality_estimate_credible_interval(CalibSets.ALPHA, 4))
#P5: Projection
# initialize a calibrated model
calibrated_model = Cls.CalibratedModel('CalibrationResults.csv')
# simulate the calibrated model
import CalibrationClasses as Cls
import CalibrationSettings as CalibSets
import SimPy.FigureSupport as Fig
import SimPy.SamplePathClasses as Path
import MultiCohortSupport as Support
import CalibrationParameterClasses as P
# initialize a calibrated model
calibrated_model_SOC = Cls.CalibratedModel(
csv_file_name='CalibrationResults.csv')
calibrated_model_NSB = Cls.CalibratedModel(
csv_file_name='CalibrationResults.csv')
# simulate the calibrated model
calibrated_model_SOC.simulate(
num_of_simulated_cohorts=CalibSets.NUM_SIM_COHORTS,
cohort_size=CalibSets.SIM_POP_SIZE,
sim_length=CalibSets.SIM_LENGTH,
diagnostic=P.Diagnostic.SOC)
calibrated_model_NSB.simulate(
num_of_simulated_cohorts=CalibSets.NUM_SIM_COHORTS,
cohort_size=CalibSets.SIM_POP_SIZE,
sim_length=CalibSets.SIM_LENGTH,
diagnostic=P.Diagnostic.NSB)
# plot the sample paths
Path.graph_sample_paths(sample_paths=calibrated_model_SOC.multiCohorts.
multiCohortOutcomes.survivalCurves,
title='Survival Curves',
x_label='Time (Week)',
import CalibrationClasses as Cls
import CalibrationSettings as Sets
import SimPy.Plots.Histogram as Hist
import SimPy.Plots.SamplePaths as Path
# initialize a calibrated model
calibrated_model = Cls.CalibratedModel(csv_file_name='CalibrationResults.csv')
# simulate the calibrated model
calibrated_model.simulate(num_of_simulated_cohorts=Sets.NUM_SIM_COHORTS,
cohort_size=Sets.SIM_POP_SIZE,
time_steps=Sets.TIME_STEPS)
# plot the sample paths
Path.plot_sample_paths(sample_paths=calibrated_model.multiCohorts.
multiCohortOutcomes.survivalCurves,
title='Survival Curves',
x_label='Time-Step (Year)',
y_label='Number Survived',
transparency=0.5)
# plot the histogram of mean survival time
Hist.plot_histogram(
data=calibrated_model.multiCohorts.multiCohortOutcomes.meanSurvivalTimes,
title='Histogram of Mean Survival Time',
x_label='Mean Survival Time (Year)',
y_label='Count',
bin_width=0.25,
x_range=[2.5, 21.5])
# create the histogram of the resampled mortality probabilities
Hist.plot_histogram(data=calibrated_model.resampledMortalityProb,
import CalibrationClasses as CalibClasses
import CalibrationSettings as CalibSets
print('Question 4')
# create a calibration object
calibration = CalibClasses.Calibration()
# sample the posterior of the mortality probability
calibration.sample_posterior()
# Estimate of mortality probability and the posterior interval
print(
'Estimate of mortality probability ({:.{prec}%} credible interval):'.
format(1 - CalibSets.ALPHA, prec=0),
calibration.get_mortality_estimate_credible_interval(CalibSets.ALPHA, 4))