def test_mock_numerical_data():
for length in [50, 100]:
for num_arrs in range(1, 3):
num_d = mock.mock_numerical_data(length=length, num_arrs=num_arrs)
assert num_d.shape[0] == num_arrs
assert num_d.shape[1] == length
for arr in num_d:
for n in arr:
assert isinstance(n, float)
assert 0 <= n <= 100000
def test_check_numerical_data():
mock_numerical = mock.mock_numerical_data(50)
# check for malformed data
# difficult to catch int arrays without running into numba type checking errors
# single dimension
with pytest.raises(ValueError):
corrupt_numerical = mock_numerical[0]
assert corrupt_numerical.ndim == 1
checks.check_numerical_data(corrupt_numerical)
# catch infinites
with pytest.raises(ValueError):
mock_numerical[0][0] = np.inf
checks.check_numerical_data(mock_numerical)
def test_metrics_to_dict(primal_graph):
# create a network layer and run some metrics
N = networks.NetworkLayerFromNX(primal_graph, distances=[500, 1000])
# check with no metrics
metrics_dict = N.metrics_to_dict()
dict_check(metrics_dict, N)
# check with centrality metrics
N.node_centrality(measures=['node_harmonic'])
metrics_dict = N.metrics_to_dict()
dict_check(metrics_dict, N)
# check with data metrics
data_dict = mock.mock_data_dict(primal_graph)
landuse_labels = mock.mock_categorical_data(len(data_dict))
numerical_data = mock.mock_numerical_data(len(data_dict))
metrics_dict = N.metrics_to_dict()
dict_check(metrics_dict, N)
def test_local_agg_time(primal_graph):
"""
Timing tests for landuse and stats aggregations
"""
if 'GITHUB_ACTIONS' in os.environ:
return
os.environ['CITYSEER_QUIET_MODE'] = '1'
# generate node and edge maps
node_uids, node_data, edge_data, node_edge_map, = graphs.graph_maps_from_nX(primal_graph)
# setup data
data_dict = mock.mock_data_dict(primal_graph, random_seed=13)
data_uids, data_map = layers.data_map_from_dict(data_dict)
data_map = data.assign_to_network(data_map, node_data, edge_data, node_edge_map, 500)
# needs a large enough beta so that distance thresholds aren't encountered
distances = np.array([np.inf])
betas = networks.beta_from_distance(distances)
qs = np.array([0, 1, 2])
mock_categorical = mock.mock_categorical_data(len(data_map))
landuse_classes, landuse_encodings = layers.encode_categorical(mock_categorical)
mock_numerical = mock.mock_numerical_data(len(data_dict), num_arrs=2, random_seed=0)
def assign_wrapper():
data.assign_to_network(data_map, node_data, edge_data, node_edge_map, 500)
# prime the function
assign_wrapper()
iters = 20000
# time and report - roughly 5.675
func_time = timeit.timeit(assign_wrapper, number=iters)
print(f'node_cent_wrapper: {func_time} for {iters} iterations')
assert func_time < 10
def landuse_agg_wrapper():
mu_data_hill, mu_data_other, ac_data, ac_data_wt = data.aggregate_landuses(node_data,
edge_data,
node_edge_map,
data_map,
distances,
betas,
mixed_use_hill_keys=np.array([0, 1]),
landuse_encodings=landuse_encodings,
qs=qs,
angular=False)
# prime the function
landuse_agg_wrapper()
iters = 20000
# time and report - roughly 10.10
func_time = timeit.timeit(landuse_agg_wrapper, number=iters)
print(f'node_cent_wrapper: {func_time} for {iters} iterations')
assert func_time < 15
def stats_agg_wrapper():
# compute
data.aggregate_stats(node_data,
edge_data,
node_edge_map,
data_map,
distances,
betas,
numerical_arrays=mock_numerical,
angular=False)
# prime the function
stats_agg_wrapper()
iters = 20000
# time and report - roughly 4.96
func_time = timeit.timeit(stats_agg_wrapper, number=iters)
print(f'segment_cent_wrapper: {func_time} for {iters} iterations')
assert func_time < 10
def test_local_aggregator_numerical_components(primal_graph):
# generate node and edge maps
node_uids, node_data, edge_data, node_edge_map = graphs.graph_maps_from_nX(primal_graph)
# setup data
data_dict = mock.mock_data_dict(primal_graph, random_seed=13)
data_uids, data_map = layers.data_map_from_dict(data_dict)
data_map = data.assign_to_network(data_map, node_data, edge_data, node_edge_map, 500)
# for debugging
# from cityseer.tools import plot
# plot.plot_graph_maps(node_uids, node_data, edge_data, data_map)
# set parameters - use a large enough distance such that simple non-weighted checks can be run for max, mean, variance
betas = np.array([0.00125])
distances = networks.distance_from_beta(betas)
mock_numerical = mock.mock_numerical_data(len(data_dict), num_arrs=2, random_seed=0)
# compute
stats_sum, stats_sum_wt, stats_mean, stats_mean_wt, stats_variance, stats_variance_wt, stats_max, stats_min = \
data.aggregate_stats(node_data,
edge_data,
node_edge_map,
data_map,
distances,
betas,
numerical_arrays=mock_numerical,
angular=False)
# non connected portions of the graph will have different stats
# used manual data plots from test_assign_to_network() to see which nodes the data points are assigned to
# connected graph is from 0 to 48 -> assigned data points are all except 5, 8, 17, 33, 48
connected_nodes_idx = list(range(49))
# and the respective data assigned to connected portion of the graph
connected_data_idx = [i for i in range(len(data_dict)) if i not in [5, 8, 9, 17, 18, 29, 33, 38, 48]]
# isolated node = 49 -> assigned no data points
# isolated nodes = 50 & 51 -> assigned data points = 17, 33
# isolated loop = 52, 53, 54, 55 -> assigned data points = 5, 8, 9, 18, 29, 38, 48
isolated_nodes_idx = [52, 53, 54, 55]
isolated_data_idx = [5, 8, 9, 18, 29, 38, 48]
for stats_idx in range(len(mock_numerical)):
for d_idx in range(len(distances)):
# max
assert np.isnan(stats_max[stats_idx, d_idx, 49])
assert np.allclose(stats_max[stats_idx, d_idx, [50, 51]], mock_numerical[stats_idx, [17, 33]].max(),
atol=0.001, rtol=0)
assert np.allclose(stats_max[stats_idx, d_idx, isolated_nodes_idx],
mock_numerical[stats_idx, isolated_data_idx].max(), atol=0.001, rtol=0)
assert np.allclose(stats_max[stats_idx, d_idx, connected_nodes_idx],
mock_numerical[stats_idx, connected_data_idx].max(), atol=0.001, rtol=0)
# min
assert np.isnan(stats_min[stats_idx, d_idx, 49])
assert np.allclose(stats_min[stats_idx, d_idx, [50, 51]], mock_numerical[stats_idx, [17, 33]].min(),
atol=0.001, rtol=0)
assert np.allclose(stats_min[stats_idx, d_idx, isolated_nodes_idx],
mock_numerical[stats_idx, isolated_data_idx].min(), atol=0.001, rtol=0)
assert np.allclose(stats_min[stats_idx, d_idx, connected_nodes_idx],
mock_numerical[stats_idx, connected_data_idx].min(), atol=0.001, rtol=0)
# sum
assert stats_sum[stats_idx, d_idx, 49] == 0
assert np.allclose(stats_sum[stats_idx, d_idx, [50, 51]],
mock_numerical[stats_idx, [17, 33]].sum(), atol=0.001, rtol=0)
assert np.allclose(stats_sum[stats_idx, d_idx, isolated_nodes_idx],
mock_numerical[stats_idx, isolated_data_idx].sum(), atol=0.001, rtol=0)
assert np.allclose(stats_sum[stats_idx, d_idx, connected_nodes_idx],
mock_numerical[stats_idx, connected_data_idx].sum(), atol=0.001, rtol=0)
# mean
assert np.isnan(stats_mean[stats_idx, d_idx, 49])
assert np.allclose(stats_mean[stats_idx, d_idx, [50, 51]], mock_numerical[stats_idx, [17, 33]].mean(),
atol=0.001, rtol=0)
assert np.allclose(stats_mean[stats_idx, d_idx, isolated_nodes_idx],
mock_numerical[stats_idx, isolated_data_idx].mean(), atol=0.001, rtol=0)
assert np.allclose(stats_mean[stats_idx, d_idx, connected_nodes_idx],
mock_numerical[stats_idx, connected_data_idx].mean(), atol=0.001, rtol=0)
# variance
assert np.isnan(stats_variance[stats_idx, d_idx, 49])
assert np.allclose(stats_variance[stats_idx, d_idx, [50, 51]], mock_numerical[stats_idx, [17, 33]].var(),
atol=0.001, rtol=0)
assert np.allclose(stats_variance[stats_idx, d_idx, isolated_nodes_idx],
mock_numerical[stats_idx, isolated_data_idx].var(), atol=0.001, rtol=0)
assert np.allclose(stats_variance[stats_idx, d_idx, connected_nodes_idx],
mock_numerical[stats_idx, connected_data_idx].var(), atol=0.001, rtol=0)
def test_compute_stats(primal_graph):
"""
Test stats component
"""
betas = np.array([0.01, 0.005])
distances = networks.distance_from_beta(betas)
# network layer
N_single = networks.NetworkLayerFromNX(primal_graph, distances=distances)
N_multi = networks.NetworkLayerFromNX(primal_graph, distances=distances)
node_map = N_multi._node_data
edge_map = N_multi._edge_data
node_edge_map = N_multi._node_edge_map
# data layer
data_dict = mock.mock_data_dict(primal_graph)
D_single = layers.DataLayerFromDict(data_dict)
D_multi = layers.DataLayerFromDict(data_dict)
# check single metrics independently against underlying for some use-cases, e.g. hill, non-hill, accessibility...
D_single.assign_to_network(N_single, max_dist=500)
D_multi.assign_to_network(N_multi, max_dist=500)
# generate some mock landuse data
mock_numeric = mock.mock_numerical_data(len(data_dict), num_arrs=2)
# generate stats
D_single.compute_stats(stats_keys='boo', stats_data_arrs=mock_numeric[0])
D_single.compute_stats(stats_keys='baa', stats_data_arrs=mock_numeric[1])
D_multi.compute_stats(stats_keys=['boo', 'baa'],
stats_data_arrs=mock_numeric)
# test against underlying method
data_map = D_single._data
stats_sum, stats_sum_wt, stats_mean, stats_mean_wt, stats_variance, stats_variance_wt, stats_max, stats_min = \
data.aggregate_stats(node_map,
edge_map,
node_edge_map,
data_map,
distances,
betas,
numerical_arrays=mock_numeric)
stats_keys = [
'max', 'min', 'sum', 'sum_weighted', 'mean', 'mean_weighted',
'variance', 'variance_weighted'
]
stats_data = [
stats_max, stats_min, stats_sum, stats_sum_wt, stats_mean,
stats_mean_wt, stats_variance, stats_variance_wt
]
for num_idx, num_label in enumerate(['boo', 'baa']):
for s_key, stats in zip(stats_keys, stats_data):
for d_idx, d_key in enumerate(distances):
# check one-at-a-time computed vs multiply computed
assert np.allclose(
N_single.metrics['stats'][num_label][s_key][d_key],
N_multi.metrics['stats'][num_label][s_key][d_key],
atol=0.001,
rtol=0,
equal_nan=True)
# check one-at-a-time against manual
assert np.allclose(
N_single.metrics['stats'][num_label][s_key][d_key],
stats[num_idx][d_idx],
atol=0.001,
rtol=0,
equal_nan=True)
# check multiply computed against manual
assert np.allclose(
N_multi.metrics['stats'][num_label][s_key][d_key],
stats[num_idx][d_idx],
atol=0.001,
rtol=0,
equal_nan=True)
# check that problematic keys and data arrays are caught
for labels, arrs, err in (
(['a'], mock_numeric, ValueError), # mismatching lengths
(['a', 'b'], None, TypeError), # missing arrays
(['a', 'b'], [], ValueError), # missing arrays
(None, mock_numeric, TypeError), # missing labels
([], mock_numeric, ValueError)): # missing labels
with pytest.raises(err):
D_multi.compute_stats(stats_keys=labels, stats_data_arrs=arrs)