def test_spatiotemporalanalysis_dataframes(self):
# test with the default constructor
sta = pls.SpatioTemporalAnalysis(self.landscape_fps)
# test that `class_metrics_df` and `landscape_metrics_df` are well
# constructed
class_metrics_df = sta.class_metrics_df
self.assertTrue(
np.all(class_metrics_df.index == pd.MultiIndex.from_product(
[sta.classes, sta.dates])))
landscape_metrics_df = sta.landscape_metrics_df
self.assertTrue(np.all(landscape_metrics_df.index == sta.dates))
# now test the same but with an analysis that only considers a
# subset of metrics and a subset of classes
sta_metrics = ['total_area', 'edge_density', 'proportion_of_landscape']
sta_classes = sta.classes[:2]
sta = pls.SpatioTemporalAnalysis(self.landscape_fps,
metrics=sta_metrics,
classes=sta_classes, dates=self.dates)
class_metrics_df = sta.class_metrics_df
self.assertTrue(
np.all(class_metrics_df.index == pd.MultiIndex.from_product(
[sta_classes, self.dates])))
landscape_metrics_df = sta.landscape_metrics_df
self.assertTrue(np.all(landscape_metrics_df.index == self.dates))
def test_spatiotemporalanalysis_dataframes(self):
# test with the default constructor
sta = pls.SpatioTemporalAnalysis(self.landscape_fps)
# test that the data frames that result from `compute_class_metrics_df`
# and `compute_landscape_metrics_df` are well constructed
class_metrics_df = sta.compute_class_metrics_df()
self.assertTrue(
np.all(class_metrics_df.index == pd.MultiIndex.from_product(
[sta.present_classes, sta.dates])))
landscape_metrics_df = sta.compute_landscape_metrics_df()
self.assertTrue(np.all(landscape_metrics_df.index == sta.dates))
# now test the same but with an analysis that only considers a
# subset of metrics and a subset of classes
metrics = ['total_area', 'edge_density', 'proportion_of_landscape']
classes = sta.present_classes[:2]
class_metrics_df = sta.compute_class_metrics_df(metrics=metrics,
classes=classes)
self.assertTrue(
np.all(class_metrics_df.index == pd.MultiIndex.from_product(
[classes, sta.dates])))
# 'proportion_of_landscape' cannot be computed at the landscape level
# (TODO: test for that elsewhere)
landscape_metrics = metrics[:2]
landscape_metrics_df = sta.compute_landscape_metrics_df(
metrics=landscape_metrics)
self.assertTrue(np.all(landscape_metrics_df.index == sta.dates))
def test_spatiotemporalanalysis_init(self):
# test that the `feature_name` is dates, and that if the `dates`
# argument is not provided when instantiating a
# `SpatioTemporalAnalysis`, the dates attribute is properly and
# automatically generated
sta = pls.SpatioTemporalAnalysis(self.landscape_fps)
self.assertEqual(sta.feature_name, 'dates')
self.assertEqual(len(sta), len(sta.dates))
def test_spatiotemporalanalysis_plot_metrics(self):
sta = pls.SpatioTemporalAnalysis(self.landscape_fps, dates=self.dates)
# test for `None` (landscape-level) and an existing class (class-level)
for class_val in [None, sta.classes[0]]:
# test that the x data of the line corresponds to the dates
self.assertTrue(
np.all(
sta.plot_metric('patch_density', class_val=class_val)
.lines[0].get_xdata() == self.dates))
def main(urban_extracts_dir, out_figure_filepath, metrics, clc_basenames,
agglomeration_slugs):
logger = logging.getLogger(__name__)
URBAN_CLASS_VAL = 1
num_rows = len(agglomeration_slugs)
num_cols = len(metrics)
figwidth, figlength = plt.rcParams['figure.figsize']
fig, axes = plt.subplots(
num_rows,
num_cols,
sharex=True,
figsize=(figwidth * num_cols, figlength * num_rows))
# extract the year code from the CLC basename, e.g., `00` from
# `g100_clc12_V18_5`
dates = [clc_basename[8:10] for clc_basename in clc_basenames]
for i, agglomeration_slug in enumerate(agglomeration_slugs):
logger.info(f'computing landscape metrics for {agglomeration_slug}')
sta = pls.SpatioTemporalAnalysis(
[
path.join(urban_extracts_dir,
f'{agglomeration_slug}-{clc_basename}.tif')
for clc_basename in clc_basenames
],
metrics=metrics,
classes=[URBAN_CLASS_VAL],
dates=dates)
for j, metric in enumerate(metrics):
sta.plot_metric(
metric,
class_val=URBAN_CLASS_VAL,
ax=axes[i, j],
metric_legend=False)
for i, agglomeration_slug in enumerate(agglomeration_slugs):
axes[i, 0].set_ylabel(agglomeration_slug.title())
for j, metric in enumerate(metrics):
axes[0, j].set_title(metric)
logger.info(f'saving figure to {out_figure_filepath}')
fig.savefig(out_figure_filepath)
