Comprehensive Analyzer Plots Example

This notebook demonstrates all available plotting methods in the DwtAnalyzer class from the tbsim package. It is adapted from the original Python script for interactive exploration.

import tbsim
import starsim as ss
import matplotlib.pyplot as plt

Build a TB simulation with dwell time analyzer

def build_tbsim(sim_pars=None):
    sim_params = dict(
        start = ss.date('2013-01-01'),
        stop = ss.date('2016-12-31'),
        rand_seed=123,
        dt=ss.days(7),
    )
    if sim_pars is not None:
        sim_params.update(sim_pars)
    pop = ss.People(n_agents=1000)
    tb_params = dict(
        beta=ss.peryear(0.025),
        init_prev=ss.bernoulli(p=0.25),
    )
    tb = tbsim.TB(tb_params)
    net = ss.RandomNet(dict(n_contacts=ss.poisson(lam=5), dur=0))
    dwell_analyzer = tbsim.DwtAnalyzer(scenario_name='comprehensive_plots_example')
    sim = ss.Sim(
        people=pop,
        networks=net,
        diseases=tb,
        pars=sim_params,
        analyzers=dwell_analyzer,
    )
    sim.pars.verbose = 0
    return sim

Run the simulation and extract the analyzer

print("Building and running TB simulation...")
sim_tb = build_tbsim()
sim_tb.run()
analyzer = sim_tb.analyzers[0]

Building and running TB simulation...

/home/runner/work/tbsim/tbsim/tbsim/analyzers.py:481: FutureWarning: The behavior of DataFrame concatenation with empty or all-NA entries is deprecated. In a future version, this will no longer exclude empty or all-NA columns when determining the result dtypes. To retain the old behavior, exclude the relevant entries before the concat operation.
  self.data = pd.concat([self.data, new_logs], ignore_index=True)

1. Network Graph

Directed graph of state transitions.

analyzer.plot('network')

<Figure size 1440x960 with 0 Axes>

2. Histograms

Histograms with KDE of dwell time distributions per state.

analyzer.plot('histogram')

3. Survival Analysis

Kaplan-Meier survival curves for dwell times.

analyzer.plot('kaplan_meier')

4. Validation

Observed dwell-time histograms for validation.

analyzer.plot('validation')

5. Loading from file

You can save the collected dwell-time data to a CSV file and later create a DwellTime instance from it.

from tbsim.analyzers import DwellTime

file_path = analyzer.save('dwelltime.csv')
print(f'Saved data file: {file_path}')

plotter = DwellTime(file_path=file_path)
plotter.plot('histogram')

Saved data file: dwelltime.csv