Example #1
0
    def test_set_outputs(self):
        model = em.PheSEIRModel()
        outputs = ['S', 'I1', 'I2', 'Incidence']
        model.set_outputs(outputs)

        with self.assertRaises(ValueError):
            outputs1 = ['S', 'E', 'I1', 'I2', 'Incidence']
            model.set_outputs(outputs1)
Example #2
0
    def test__init__(self):
        model = em.PheSEIRModel()

        self.assertEqual(model._output_names,
                         ['S', 'E1', 'E2', 'I1', 'I2', 'R', 'Incidence'])
        self.assertEqual(
            model._parameter_names,
            ['S0', 'E10', 'E20', 'I10', 'I20', 'R0', 'beta', 'kappa', 'gamma'])
        self.assertEqual(model._n_outputs, 7)
        self.assertEqual(model._n_parameters, 9)
Example #3
0
    def test_samples_positive_tests(self):
        model = em.PheSEIRModel()

        # Populate the model
        regions = ['London', 'Cornwall']
        age_groups = ['0-10', '10-25']

        # Initial state of the system
        contact_data_matrix_0 = np.array([[1, 0], [0, 3]])
        contact_data_matrix_1 = np.array([[10, 5.2], [0, 3]])

        region_data_matrix_0_0 = np.array([[0.5, 0], [0, 6]])
        region_data_matrix_0_1 = np.array([[1, 10], [1, 0]])
        region_data_matrix_1_0 = np.array([[0.5, 1.2], [0.29, 6]])
        region_data_matrix_1_1 = np.array([[0.85, 1], [0.9, 6]])

        susceptibles = [[5, 6], [7, 8]]
        dI = 4

        contacts_0 = em.ContactMatrix(age_groups, contact_data_matrix_0)
        contacts_1 = em.ContactMatrix(age_groups, contact_data_matrix_1)
        regional_0_0 = em.RegionMatrix(regions[0], age_groups,
                                       region_data_matrix_0_0)
        regional_0_1 = em.RegionMatrix(regions[1], age_groups,
                                       region_data_matrix_0_1)
        regional_1_0 = em.RegionMatrix(regions[0], age_groups,
                                       region_data_matrix_1_0)
        regional_1_1 = em.RegionMatrix(regions[1], age_groups,
                                       region_data_matrix_1_1)

        # Matrices contact
        matrices_contact = [contacts_0, contacts_1]
        time_changes_contact = [1, 3]
        matrices_region = [[regional_0_0, regional_0_1],
                           [regional_1_0, regional_1_1]]
        time_changes_region = [1, 2]

        model.set_regions(regions)
        model.read_contact_data(matrices_contact, time_changes_contact)
        model.read_regional_data(matrices_region, time_changes_region)

        initial_r = [0.5, 1]

        parameters = [
            initial_r, 1, susceptibles, [[0, 0], [0, 0]], [[0.1, 0.2], [0, 0]],
            [[0, 0], [0, 0]], [[0, 0], [0, 0]], [[0, 0], [0, 0]],
            [[1] * 2, [1] * 2], 4, dI, 0.5, 'my-solver'
        ]

        times = [1, 2]

        output = model.simulate(list(deepflatten(parameters, ignore=str)),
                                times)

        tests = [[20, 30], [10, 0]]
        sens = 0.9
        spec = 0.1

        self.assertEqual(
            model.samples_positive_tests(output, tests[0], sens, spec,
                                         0).shape, (len(age_groups), ))

        with self.assertRaises(TypeError):
            model.samples_positive_tests(output, tests[0], sens, spec, '1')

        with self.assertRaises(ValueError):
            model.samples_positive_tests(output, tests[0], sens, spec, -1)

        with self.assertRaises(ValueError):
            model.samples_positive_tests(output, tests[0], sens, spec, 3)
Example #4
0
    def test_check_positives_format(self):
        model = em.PheSEIRModel()

        # Populate the model
        regions = ['London', 'Cornwall']
        age_groups = ['0-10', '10-25']

        # Initial state of the system
        contact_data_matrix_0 = np.array([[1, 0], [0, 3]])
        contact_data_matrix_1 = np.array([[10, 5.2], [0, 3]])

        region_data_matrix_0_0 = np.array([[0.5, 0], [0, 6]])
        region_data_matrix_0_1 = np.array([[1, 10], [1, 0]])
        region_data_matrix_1_0 = np.array([[0.5, 1.2], [0.29, 6]])
        region_data_matrix_1_1 = np.array([[0.85, 1], [0.9, 6]])

        susceptibles = [[5, 6], [7, 8]]
        dI = 4

        contacts_0 = em.ContactMatrix(age_groups, contact_data_matrix_0)
        contacts_1 = em.ContactMatrix(age_groups, contact_data_matrix_1)
        regional_0_0 = em.RegionMatrix(regions[0], age_groups,
                                       region_data_matrix_0_0)
        regional_0_1 = em.RegionMatrix(regions[1], age_groups,
                                       region_data_matrix_0_1)
        regional_1_0 = em.RegionMatrix(regions[0], age_groups,
                                       region_data_matrix_1_0)
        regional_1_1 = em.RegionMatrix(regions[1], age_groups,
                                       region_data_matrix_1_1)

        # Matrices contact
        matrices_contact = [contacts_0, contacts_1]
        time_changes_contact = [1, 3]
        matrices_region = [[regional_0_0, regional_0_1],
                           [regional_1_0, regional_1_1]]
        time_changes_region = [1, 2]

        model.set_regions(regions)
        model.read_contact_data(matrices_contact, time_changes_contact)
        model.read_regional_data(matrices_region, time_changes_region)

        initial_r = [0.5, 1]

        parameters = [
            initial_r, 1, susceptibles, [[0, 0], [0, 0]], [[0.1, 0.2], [0, 0]],
            [[0, 0], [0, 0]], [[0, 0], [0, 0]], [[0, 0], [0, 0]],
            [[1] * 2, [1] * 2], 4, dI, 0.5, 'my-solver'
        ]

        times = [1, 2]

        output = model.simulate(list(deepflatten(parameters, ignore=str)),
                                times)

        tests = [[20, 30], [10, 0]]
        sens = 0.9
        spec = 0.1

        with self.assertRaises(ValueError):
            output1 = np.array([5, 6, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0])

            model.check_positives_format(output1, tests, sens, spec)

        with self.assertRaises(ValueError):
            output1 = np.array([[5, 6, 0, 0, 0, 0, 0, 0, 0, 0],
                                [5, 6, 0, 0, 0, 0, 0, 0, 0, 0]])

            model.check_positives_format(output1, tests, sens, spec)

        with self.assertRaises(ValueError):
            output1 = np.array([[5, 6, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
                                [5, 6, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
                                [5, 6, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]])

            model.check_positives_format(output1, tests, sens, spec)

        with self.assertRaises(TypeError):
            output1 = np.array([['5', 6, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
                                [5, 6, '0', 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]])

            model.check_positives_format(output1, tests, sens, spec)

        with self.assertRaises(ValueError):
            tests1 = 100

            model.check_positives_format(output, tests1, sens, spec)

        with self.assertRaises(ValueError):
            tests1 = np.array([2, 50])

            model.check_positives_format(output, tests1, sens, spec)

        with self.assertRaises(ValueError):
            tests1 = np.array([[20, 30, 1], [10, 0, 0]])

            model.check_positives_format(output, tests1, sens, spec)

        with self.assertRaises(TypeError):
            tests1 = np.array([[20, '30'], [10, 0]])

            model.check_positives_format(output, tests1, sens, spec)

        with self.assertRaises(ValueError):
            tests1 = np.array([[-1, 50], [10, 0]])

            model.check_positives_format(output, tests1, sens, spec)

        with self.assertRaises(TypeError):
            model.check_positives_format(output, tests, '0.9', spec)

        with self.assertRaises(ValueError):
            model.check_positives_format(output, tests, -0.2, spec)

        with self.assertRaises(ValueError):
            model.check_positives_format(output, tests, 1.2, spec)

        with self.assertRaises(TypeError):
            model.check_positives_format(output, tests, sens, '0.1')

        with self.assertRaises(ValueError):
            model.check_positives_format(output, tests, sens, -0.1)

        with self.assertRaises(ValueError):
            model.check_positives_format(output, tests, sens, 1.2)
Example #5
0
    def test_simulate(self):
        model = em.PheSEIRModel()

        # Populate the model
        regions = ['London', 'Cornwall']
        age_groups = ['0-10', '10-25']

        # Initial state of the system
        contact_data_matrix_0 = np.array([[1, 0], [0, 3]])
        contact_data_matrix_1 = np.array([[10, 5.2], [0, 3]])

        region_data_matrix_0_0 = np.array([[0.5, 0], [0, 6]])
        region_data_matrix_0_1 = np.array([[1, 10], [1, 0]])
        region_data_matrix_1_0 = np.array([[0.5, 1.2], [0.29, 6]])
        region_data_matrix_1_1 = np.array([[0.85, 1], [0.9, 6]])

        susceptibles = [[5, 6], [7, 8]]
        dI = 4

        contacts_0 = em.ContactMatrix(age_groups, contact_data_matrix_0)
        contacts_1 = em.ContactMatrix(age_groups, contact_data_matrix_1)
        regional_0_0 = em.RegionMatrix(regions[0], age_groups,
                                       region_data_matrix_0_0)
        regional_0_1 = em.RegionMatrix(regions[1], age_groups,
                                       region_data_matrix_0_1)
        regional_1_0 = em.RegionMatrix(regions[0], age_groups,
                                       region_data_matrix_1_0)
        regional_1_1 = em.RegionMatrix(regions[1], age_groups,
                                       region_data_matrix_1_1)

        # Matrices contact
        matrices_contact = [contacts_0, contacts_1]
        time_changes_contact = [1, 3]
        matrices_region = [[regional_0_0, regional_0_1],
                           [regional_1_0, regional_1_1]]
        time_changes_region = [1, 2]

        model.set_regions(regions)
        model.read_contact_data(matrices_contact, time_changes_contact)
        model.read_regional_data(matrices_region, time_changes_region)

        initial_r = [0.5, 1]

        parameters = [
            initial_r, 2, susceptibles, [[0, 0], [0, 0]], [[0, 0], [0, 0]],
            [[0, 0], [0, 0]], [[0, 0], [0, 0]], [[0, 0], [0, 0]],
            [[1] * 2, [1] * 2], 4, dI, 0.5, 'RK45'
        ]

        times = [1, 2]

        output_my_solver = model.simulate(
            list(deepflatten(parameters, ignore=str)), times)

        npt.assert_almost_equal(
            output_my_solver,
            np.array([[7, 8, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
                      [7, 8, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]]),
            decimal=3)

        parameters[-1] = 'my-solver'

        output_scipy_solver = model.simulate(
            list(deepflatten(parameters, ignore=str)), times)

        npt.assert_almost_equal(
            output_scipy_solver,
            np.array([[7, 8, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
                      [7, 8, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]]),
            decimal=3)

        parameters[-1] = 'my-solver'

        with self.assertRaises(TypeError):
            model.simulate(list(deepflatten(parameters, ignore=str)), '0')

        with self.assertRaises(TypeError):
            model.simulate(list(deepflatten(parameters, ignore=str)), ['1', 2])

        with self.assertRaises(ValueError):
            model.simulate(list(deepflatten(parameters, ignore=str)), [0, 1])

        with self.assertRaises(TypeError):
            model.simulate('parameters', times)

        with self.assertRaises(ValueError):
            model.simulate([0], times)

        with self.assertRaises(TypeError):
            parameters1 = [
                initial_r, 0.5, susceptibles, [[0, 0], [0, 0]], [[0, 0],
                                                                 [0, 0]],
                [[0, 0], [0, 0]], [[0, 0], [0, 0]], [[0, 0], [0, 0]],
                [[1] * 2, [1] * 2], 4, dI, 0.005, 'my-solver'
            ]

            model.simulate(list(deepflatten(parameters1, ignore=(str, float))),
                           times)

        with self.assertRaises(ValueError):
            parameters1 = [
                initial_r, 0, susceptibles, [[0, 0], [0, 0]], [[0, 0], [0, 0]],
                [[0, 0], [0, 0]], [[0, 0], [0, 0]], [[0, 0], [0, 0]],
                [[1] * 2, [1] * 2], 4, dI, 0.005, 'my-solver'
            ]

            model.simulate(list(deepflatten(parameters1, ignore=str)), times)

        with self.assertRaises(ValueError):
            parameters1 = [
                initial_r, 3, susceptibles, [[0, 0], [0, 0]], [[0, 0], [0, 0]],
                [[0, 0], [0, 0]], [[0, 0], [0, 0]], [[0, 0], [0, 0]],
                [[1] * 2, [1] * 2], 4, dI, 0.005, 'my-solver'
            ]

            model.simulate(list(deepflatten(parameters1, ignore=str)), times)

        with self.assertRaises(ValueError):
            susceptibles1 = [5, 6]

            parameters1 = [
                initial_r, 1, susceptibles1, [[0, 0], [0, 0]], [[0, 0], [0,
                                                                         0]],
                [[0, 0], [0, 0]], [[0, 0], [0, 0]], [[0, 0], [0, 0]],
                [[1] * 2, [1] * 2], 4, dI, 0.005, 'my-solver'
            ]

            model.simulate(list(deepflatten(parameters1, ignore=str)), times)

        with self.assertRaises(ValueError):
            parameters1 = [
                initial_r, 1, susceptibles, [[0, 0], [0, 0]],
                [[0, 0], [0, 0], [0, 0]], [[0, 0], [0, 0]], [[0, 0], [0, 0]],
                [[0, 0], [0, 0]], [[1] * 2, [1] * 2], 4, dI, 0.5, 'my-solver'
            ]

            model.simulate(list(deepflatten(parameters1, ignore=str)), times)

        with self.assertRaises(ValueError):
            parameters1 = [
                initial_r, 1, susceptibles, [[0, 0], [0, 0]], [[0, 0], [0, 0]],
                [[0, 0], [0, 0]], [[0, 0, 0], [0, 0, 0]], [[0, 0], [0, 0]],
                [[1] * 2, [1] * 2], 4, dI, 0.005, 'my-solver'
            ]

            model.simulate(list(deepflatten(parameters1, ignore=str)), times)

        with self.assertRaises(ValueError):
            parameters1 = [
                initial_r, 1, susceptibles, [[0, 0], [0, 0]], [[0, 0], [0, 0]],
                [[0, 0], [0, 0]], [[0, 0], [0, 0]], [[0, 0], [0, 0]],
                [[[1] * 2, [1] * 2], [[1] * 2, [1] * 2]], 4, dI, 0.005,
                'my-solver'
            ]

            model.simulate(list(deepflatten(parameters1, ignore=str)), times)

        with self.assertRaises(ValueError):
            parameters1 = [
                initial_r, 1, susceptibles, [[0, 0], [0, 0]], [[0, 0], [0, 0]],
                [[0, 0], [0, 0]], [[0, 0], [0, 0]], [[0, 0], [0, 0]],
                [[1] * 2, [1] * 2, [1] * 2], 4, dI, 0.005, 'my-solver'
            ]

            model.simulate(list(deepflatten(parameters1, ignore=str)), times)

        with self.assertRaises(ValueError):
            parameters1 = [
                initial_r, 1, susceptibles, [[0, 0], [0, 0]], [[0, 0], [0, 0]],
                [[0, 0], [0, 0]], [[0, 0], [0, 0]], [[0, 0], [0, 0]],
                [[1] * 4, [1] * 4], 4, dI, 0.005, 'my-solver'
            ]

            model.simulate(list(deepflatten(parameters1, ignore=str)), times)

        with self.assertRaises(TypeError):
            parameters1 = [
                initial_r, 1, susceptibles, [[0, 0], [0, 0]], [[0, 0], [0, 0]],
                [[0, 0], [0, 0]], [[0, 0], [0, 0]], [[0, 0], [0, 0]],
                [[1] * 2, [1] * 2], '4', dI, 0.005, 'my-solver'
            ]

            model.simulate(list(deepflatten(parameters1, ignore=str)), times)

        with self.assertRaises(ValueError):
            parameters1 = [
                initial_r, 1, susceptibles, [[0, 0], [0, 0]], [[0, 0], [0, 0]],
                [[0, 0], [0, 0]], [[0, 0], [0, 0]], [[0, 0], [0, 0]],
                [[1] * 2, [1] * 2], -1, dI, 0.005, 'my-solver'
            ]

            model.simulate(list(deepflatten(parameters1, ignore=str)), times)

        with self.assertRaises(TypeError):
            parameters1 = [
                initial_r, 1, susceptibles, [[0, 0], [0, 0]], [[0, 0], [0, 0]],
                [[0, 0], [0, 0]], [[0, 0], [0, 0]], [[0, 0], [0, 0]],
                [[1] * 2, [1] * 2], 4, '4', 0.005, 'my-solver'
            ]

            model.simulate(list(deepflatten(parameters1, ignore=str)), times)

        with self.assertRaises(ValueError):
            parameters1 = [
                initial_r, 1, susceptibles, [[0, 0], [0, 0]], [[0, 0], [0, 0]],
                [[0, 0], [0, 0]], [[0, 0], [0, 0]], [[0, 0], [0, 0]],
                [[1] * 2, [1] * 2], 4, 0, 0.005, 'my-solver'
            ]

            model.simulate(list(deepflatten(parameters1, ignore=str)), times)

        with self.assertRaises(TypeError):
            parameters1 = [
                initial_r, 1, susceptibles, [[0, 0], [0, 0]], [[0, 0], [0, 0]],
                [[0, 0], [0, 0]], [[0, 0], [0, 0]], [[0, 0], [0, 0]],
                [[1] * 2, [1] * 2], 4, dI, '0.005', 'my-solver'
            ]

            model.simulate(list(deepflatten(parameters1, ignore=str)), times)

        with self.assertRaises(ValueError):
            parameters1 = [
                initial_r, 1, susceptibles, [[0, 0], [0, 0]], [[0, 0], [0, 0]],
                [[0, 0], [0, 0]], [[0, 0], [0, 0]], [[0, 0], [0, 0]],
                [[1] * 2, [1] * 2], 4, dI, 0, 'my-solver'
            ]

            model.simulate(list(deepflatten(parameters1, ignore=str)), times)

        with self.assertRaises(TypeError):
            parameters1 = [
                initial_r, 1, susceptibles, [[0, 0], [0, 0]], [[0, 0], [0, 0]],
                [[0, 0], [0, 0]], [[0, 0], [0, 0]], [[0, 0], [0, 0]],
                [[1] * 2, [1] * 2], 4, dI, 0.5, 3
            ]

            model.simulate(list(deepflatten(parameters1, ignore=str)), times)

        with self.assertRaises(ValueError):
            parameters1 = [
                initial_r, 1, susceptibles, [[0, 0], [0, 0]], [[0, 0], [0, 0]],
                [[0, 0], [0, 0]], [[0, 0], [0, 0]], [[0, 0], [0, 0]],
                [[1] * 2, [1] * 2], 4, dI, 0.5, 'my-solver2'
            ]

            model.simulate(list(deepflatten(parameters1, ignore=str)), times)
Example #6
0
    def test_samples_deaths(self):
        model = em.PheSEIRModel()

        # Populate the model
        regions = ['London', 'Cornwall']
        age_groups = ['0-10', '10-25']

        # Initial state of the system
        contact_data_matrix_0 = np.array([[1, 0], [0, 3]])
        contact_data_matrix_1 = np.array([[10, 5.2], [0, 3]])

        region_data_matrix_0_0 = np.array([[0.5, 0], [0, 6]])
        region_data_matrix_0_1 = np.array([[1, 10], [1, 0]])
        region_data_matrix_1_0 = np.array([[0.5, 1.2], [0.29, 6]])
        region_data_matrix_1_1 = np.array([[0.85, 1], [0.9, 6]])

        susceptibles = [[5, 6], [7, 8]]
        dI = 4

        contacts_0 = em.ContactMatrix(age_groups, contact_data_matrix_0)
        contacts_1 = em.ContactMatrix(age_groups, contact_data_matrix_1)
        regional_0_0 = em.RegionMatrix(regions[0], age_groups,
                                       region_data_matrix_0_0)
        regional_0_1 = em.RegionMatrix(regions[1], age_groups,
                                       region_data_matrix_0_1)
        regional_1_0 = em.RegionMatrix(regions[0], age_groups,
                                       region_data_matrix_1_0)
        regional_1_1 = em.RegionMatrix(regions[1], age_groups,
                                       region_data_matrix_1_1)

        # Matrices contact
        matrices_contact = [contacts_0, contacts_1]
        time_changes_contact = [1, 3]
        matrices_region = [[regional_0_0, regional_0_1],
                           [regional_1_0, regional_1_1]]
        time_changes_region = [1, 2]

        model.set_regions(regions)
        model.read_contact_data(matrices_contact, time_changes_contact)
        model.read_regional_data(matrices_region, time_changes_region)

        initial_r = [0.5, 1]

        parameters = [
            initial_r, 1, susceptibles, [[0, 0], [0, 0]], [[0.1, 0.2], [0, 0]],
            [[0, 0], [0, 0]], [[0, 0], [0, 0]], [[0, 0], [0, 0]],
            [[1] * 60, [1] * 60], 4, dI, 0.5, 'my-solver'
        ]

        times = np.arange(1, 61).tolist()

        output = model.simulate(list(deepflatten(parameters, ignore=str)),
                                times)

        new_infections = model.new_infections(output)

        fatality_ratio = [0.1, 0.5]

        td_mean = 15.0
        td_var = 12.1**2
        theta = td_var / td_mean
        k = td_mean / theta
        time_to_death = gamma(k, scale=theta).pdf(np.arange(1, 60)).tolist()

        self.assertEqual(
            model.samples_deaths(new_infections, fatality_ratio, time_to_death,
                                 0.5, 41).shape, (len(age_groups), ))

        self.assertEqual(
            model.samples_deaths(new_infections, fatality_ratio, time_to_death,
                                 0.5, 1).shape, (len(age_groups), ))

        with self.assertRaises(ValueError):
            model.samples_deaths(new_infections, fatality_ratio, time_to_death,
                                 0.5, -1)

        with self.assertRaises(TypeError):
            model.samples_deaths(new_infections, fatality_ratio, time_to_death,
                                 0.5, '1')

        with self.assertRaises(ValueError):
            model.samples_deaths(new_infections, fatality_ratio, time_to_death,
                                 0.5, 62)
Example #7
0
    def test_set_age_groups(self):
        model = em.PheSEIRModel()
        age_groups = ['0-10', '10-20']
        model.set_age_groups(age_groups)

        self.assertEqual(model.age_groups_names(), ['0-10', '10-20'])
Example #8
0
    def test_set_regions(self):
        model = em.PheSEIRModel()
        regions = ['London', 'Cornwall']
        model.set_regions(regions)

        self.assertEqual(model.region_names(), ['London', 'Cornwall'])
Example #9
0
    def test_check_death_format(self):
        model = em.PheSEIRModel()

        # Populate the model
        regions = ['London', 'Cornwall']
        age_groups = ['0-10', '10-25']

        # Initial state of the system
        contact_data_matrix_0 = np.array([[1, 0], [0, 3]])
        contact_data_matrix_1 = np.array([[10, 5.2], [0, 3]])

        region_data_matrix_0_0 = np.array([[0.5, 0], [0, 6]])
        region_data_matrix_0_1 = np.array([[1, 10], [1, 0]])
        region_data_matrix_1_0 = np.array([[0.5, 1.2], [0.29, 6]])
        region_data_matrix_1_1 = np.array([[0.85, 1], [0.9, 6]])

        susceptibles = [[5, 6], [7, 8]]
        dI = 4

        contacts_0 = em.ContactMatrix(age_groups, contact_data_matrix_0)
        contacts_1 = em.ContactMatrix(age_groups, contact_data_matrix_1)
        regional_0_0 = em.RegionMatrix(regions[0], age_groups,
                                       region_data_matrix_0_0)
        regional_0_1 = em.RegionMatrix(regions[1], age_groups,
                                       region_data_matrix_0_1)
        regional_1_0 = em.RegionMatrix(regions[0], age_groups,
                                       region_data_matrix_1_0)
        regional_1_1 = em.RegionMatrix(regions[1], age_groups,
                                       region_data_matrix_1_1)

        # Matrices contact
        matrices_contact = [contacts_0, contacts_1]
        time_changes_contact = [1, 3]
        matrices_region = [[regional_0_0, regional_0_1],
                           [regional_1_0, regional_1_1]]
        time_changes_region = [1, 2]

        model.set_regions(regions)
        model.read_contact_data(matrices_contact, time_changes_contact)
        model.read_regional_data(matrices_region, time_changes_region)

        initial_r = [0.5, 1]

        parameters = [
            initial_r, 1, susceptibles, [[0, 0], [0, 0]], [[0.1, 0.2], [0, 0]],
            [[0, 0], [0, 0]], [[0, 0], [0, 0]], [[0, 0], [0, 0]],
            [[1] * 2, [1] * 2], 4, dI, 0.5, 'my-solver'
        ]

        times = [1, 2]

        output = model.simulate(list(deepflatten(parameters, ignore=str)),
                                times)

        new_infections = model.new_infections(output)

        fatality_ratio = [0.1, 0.5]
        time_to_death = [0.5, 0.5]

        with self.assertRaises(TypeError):
            model.check_death_format(new_infections, fatality_ratio,
                                     time_to_death, '0.5')

        with self.assertRaises(ValueError):
            model.check_death_format(new_infections, fatality_ratio,
                                     time_to_death, -2)

        with self.assertRaises(ValueError):
            new_infections1 = \
                np.array([5, 6])

            model.check_death_format(new_infections1, fatality_ratio,
                                     time_to_death, 0.5)

        with self.assertRaises(ValueError):
            new_infections1 = np.array([[5, 6, 0, 0], [5, 6, 0, 0]])

            model.check_death_format(new_infections1, fatality_ratio,
                                     time_to_death, 0.5)

        with self.assertRaises(ValueError):
            new_infections1 = np.array([[5, 6], [5, 6], [5, 6]])

            model.check_death_format(new_infections1, fatality_ratio,
                                     time_to_death, 0.5)

        with self.assertRaises(TypeError):
            new_infections1 = np.array([['5', 6], [5, '0']])

            model.check_death_format(new_infections1, fatality_ratio,
                                     time_to_death, 0.5)

        with self.assertRaises(ValueError):
            fatality_ratio1 = 0

            model.check_death_format(new_infections, fatality_ratio1,
                                     time_to_death, 0.5)

        with self.assertRaises(ValueError):
            fatality_ratio1 = np.array([0.1, 0.5, 0.1])

            model.check_death_format(new_infections, fatality_ratio1,
                                     time_to_death, 0.5)

        with self.assertRaises(TypeError):
            fatality_ratio1 = np.array([0.1, '0.5'])

            model.check_death_format(new_infections, fatality_ratio1,
                                     time_to_death, 0.5)

        with self.assertRaises(ValueError):
            fatality_ratio1 = np.array([-0.1, 0.5])

            model.check_death_format(new_infections, fatality_ratio1,
                                     time_to_death, 0.5)

        with self.assertRaises(ValueError):
            fatality_ratio1 = np.array([0.1, 1.5])

            model.check_death_format(new_infections, fatality_ratio1,
                                     time_to_death, 0.5)

        with self.assertRaises(ValueError):
            time_to_death1 = np.array([[0.5], [0.5]])

            model.check_death_format(new_infections, fatality_ratio,
                                     time_to_death1, 0.5)

        with self.assertRaises(ValueError):
            time_to_death1 = np.array([0.5, 0.5, 0.15])

            model.check_death_format(new_infections, fatality_ratio,
                                     time_to_death1, 0.5)

        with self.assertRaises(TypeError):
            time_to_death1 = np.array(['0.1', 0.5])

            model.check_death_format(new_infections, fatality_ratio,
                                     time_to_death1, 0.5)

        with self.assertRaises(ValueError):
            time_to_death1 = np.array([-0.1, 0.5])

            model.check_death_format(new_infections, fatality_ratio,
                                     time_to_death1, 0.5)

        with self.assertRaises(ValueError):
            time_to_death1 = np.array([0.5, 1.1])

            model.check_death_format(new_infections, fatality_ratio,
                                     time_to_death1, 0.5)
Example #10
0
 def test_parameter_names(self):
     model = em.PheSEIRModel()
     self.assertEqual(
         model.parameter_names(),
         ['S0', 'E10', 'E20', 'I10', 'I20', 'R0', 'beta', 'kappa', 'gamma'])
Example #11
0
 def test_output_names(self):
     model = em.PheSEIRModel()
     self.assertEqual(model.output_names(),
                      ['S', 'E1', 'E2', 'I1', 'I2', 'R', 'Incidence'])
Example #12
0
 def test_n_parameters(self):
     model = em.PheSEIRModel()
     self.assertEqual(model.n_parameters(), 9)
Example #13
0
    def test_new_infections(self):
        model = em.PheSEIRModel()

        # Populate the model
        regions = ['London', 'Cornwall']
        age_groups = ['0-10', '10-25']

        # Initial state of the system
        contact_data_matrix_0 = np.array([[1, 0], [0, 3]])
        contact_data_matrix_1 = np.array([[10, 5.2], [0, 3]])

        region_data_matrix_0_0 = np.array([[0.5, 0], [0, 6]])
        region_data_matrix_0_1 = np.array([[1, 10], [1, 0]])
        region_data_matrix_1_0 = np.array([[0.5, 1.2], [0.29, 6]])
        region_data_matrix_1_1 = np.array([[0.85, 1], [0.9, 6]])

        susceptibles = [[5, 6], [7, 8]]
        dI = 4

        contacts_0 = em.ContactMatrix(age_groups, contact_data_matrix_0)
        contacts_1 = em.ContactMatrix(age_groups, contact_data_matrix_1)
        regional_0_0 = em.RegionMatrix(regions[0], age_groups,
                                       region_data_matrix_0_0)
        regional_0_1 = em.RegionMatrix(regions[1], age_groups,
                                       region_data_matrix_0_1)
        regional_1_0 = em.RegionMatrix(regions[0], age_groups,
                                       region_data_matrix_1_0)
        regional_1_1 = em.RegionMatrix(regions[1], age_groups,
                                       region_data_matrix_1_1)

        # Matrices contact
        matrices_contact = [contacts_0, contacts_1]
        time_changes_contact = [1, 3]
        matrices_region = [[regional_0_0, regional_0_1],
                           [regional_1_0, regional_1_1]]
        time_changes_region = [1, 2]

        model.set_regions(regions)
        model.read_contact_data(matrices_contact, time_changes_contact)
        model.read_regional_data(matrices_region, time_changes_region)

        initial_r = [0.5, 1]

        parameters = [
            initial_r, 1, susceptibles, [[0, 0], [0, 0]], [[0, 0], [0, 0]],
            [[0, 0], [0, 0]], [[0, 0], [0, 0]], [[0, 0], [0, 0]],
            [[1] * 2, [1] * 2], 4, dI, 0.5, 'my-solver'
        ]

        times = [1, 2]

        output = model.simulate(list(deepflatten(parameters, ignore=str)),
                                times)

        npt.assert_array_equal(model.new_infections(output),
                               np.array([[0, 0], [0, 0]]))

        with self.assertRaises(ValueError):
            output1 = np.array([5, 6, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0])
            model.new_infections(output1)

        with self.assertRaises(ValueError):
            output1 = np.array([[5, 6, 0, 0, 0, 0, 0, 0, 0, 0],
                                [5, 6, 0, 0, 0, 0, 0, 0, 0, 0]])
            model.new_infections(output1)

        with self.assertRaises(ValueError):
            output1 = np.array([[5, 6, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
                                [5, 6, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
                                [5, 6, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]])
            model.new_infections(output1)

        with self.assertRaises(TypeError):
            output1 = np.array([['5', 6, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
                                [5, 6, '0', 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]])
            model.new_infections(output1)
Example #14
0
 def test_n_outputs(self):
     model = em.PheSEIRModel()
     self.assertEqual(model.n_outputs(), 7)