def test_distribution_plot(artworks_df, artworks_summary):
def mock_render(fig):
# check that this draws a histogram
assert len(fig.axes[0].patches) > 0
explorer = Explorer(artworks_summary, plot_renderer=mock_render)
explorer.distribution_plot("Height (cm)")
def test_cdf_plot_non_numeric(artworks_summary):
def mock_render(fig):
assert False
explorer = Explorer(artworks_summary, plot_renderer=mock_render)
with pytest.raises(ValueError):
explorer.cdf_plot("Nationality")
def test_distribution_plot_bins(artworks_df, artworks_summary):
Nbins = 13
def mock_render(fig):
# check that this draws a histogram with Nbins bars
assert len(fig.axes[0].patches) == Nbins
explorer = Explorer(artworks_summary, plot_renderer=mock_render)
explorer.distribution_plot("Height (cm)", bins=Nbins)
def test_pairwise_density_plot_not_numeric(artworks_df, artworks_summary):
plt.cla()
def mock_render(fig):
assert False
explorer = Explorer(artworks_summary, plot_renderer=mock_render)
with pytest.raises(ValueError):
explorer.pairwise_density_plot("Diameter (cm)", "Nationality")
def test_cdf_plot_log_transformed(artworks_df, artworks_summary):
plt.cla()
def mock_render(fig):
ax = fig.axes[0]
assert len(ax.lines) == 1
assert ax.get_xaxis().get_scale() == "log"
explorer = Explorer(artworks_summary, plot_renderer=mock_render)
explorer.cdf_plot("Width (cm)")
def test_pairwise_density_plot(artworks_df, artworks_summary):
plt.cla()
def mock_render(fig):
# currently pairwise_density_plot returns a plotly figure
assert len(fig["data"]) == 1
data = fig["data"][0]
assert data["type"] == "heatmap"
explorer = Explorer(artworks_summary, plot_renderer=mock_render)
explorer.pairwise_density_plot("Width (cm)", "Height (cm)")
def test_correlation_plot_include(artworks_df, artworks_summary):
plt.cla()
def mock_render(fig):
assert len(fig["data"]) == 1
data = fig["data"][0]
assert data["type"] == "heatmap"
assert set(data["y"]) == set(data["x"]) == {"Height (cm)"}
explorer = Explorer(artworks_summary, plot_renderer=mock_render)
explorer.correlation_plot(include=["Height (cm)"])
def test_pairwise_density_plot_both_categorical(artworks_df, artworks_summary):
plt.cla()
def mock_render(fig):
# currently pairwise_density_plot returns a plotly figure
assert len(fig["data"]) == 1
data = fig["data"][0]
assert data["type"] == "heatmap"
explorer = Explorer(artworks_summary, plot_renderer=mock_render)
explorer.pairwise_density_plot("Nationality", "Gender")
def test_pairwise_density_plot(artworks_df, artworks_summary):
plt.cla()
def mock_render(fig):
# currently pairwise_density_plot returns a plotly figure
assert len(fig['data']) == 1
data = fig['data'][0]
assert data['type'] == 'heatmap'
explorer = Explorer(artworks_summary, plot_renderer=mock_render)
explorer.pairwise_density_plot('Width (cm)', 'Height (cm)')
def test_correlation_plot_include(artworks_df, artworks_summary):
plt.cla()
def mock_render(fig):
assert len(fig['data']) == 1
data = fig['data'][0]
assert data['type'] == 'heatmap'
assert set(data['y']) == set(data['x']) == {'Height (cm)'}
explorer = Explorer(artworks_summary, plot_renderer=mock_render)
explorer.correlation_plot(include=['Height (cm)'])
def test_pairwise_density_plot_both_categorical(artworks_df, artworks_summary):
plt.cla()
def mock_render(fig):
# currently pairwise_density_plot returns a plotly figure
assert len(fig['data']) == 1
data = fig['data'][0]
assert data['type'] == 'heatmap'
explorer = Explorer(artworks_summary, plot_renderer=mock_render)
explorer.pairwise_density_plot('Nationality', 'Gender')
def test_correlation_plot_exclude(artworks_df, artworks_summary):
plt.cla()
def mock_render(fig):
assert len(fig["data"]) == 1
data = fig["data"][0]
assert data["type"] == "heatmap"
expected_columns = {"Height (cm)", "Depth (cm)"}
assert set(data["y"]) == set(data["x"]) == expected_columns
explorer = Explorer(artworks_summary, plot_renderer=mock_render)
explorer.correlation_plot(exclude=["Width (cm)"])
def test_correlation_plot_exclude(artworks_df, artworks_summary):
plt.cla()
def mock_render(fig):
assert len(fig['data']) == 1
data = fig['data'][0]
assert data['type'] == 'heatmap'
expected_columns = {'Height (cm)', 'Depth (cm)'}
assert set(data['y']) == set(data['x']) == expected_columns
explorer = Explorer(artworks_summary, plot_renderer=mock_render)
explorer.correlation_plot(exclude=['Width (cm)'])
def test_correlation_plot_annotations(artworks_df, artworks_summary):
plt.cla()
def mock_render(fig):
assert len(fig["data"]) == 1
corr = [item for row in fig["data"][0]["z"] for item in row]
labels = [l["text"] for l in fig["layout"]["annotations"]]
for c, l in zip(corr, labels):
assert "{:.2g}".format(c) == l
explorer = Explorer(artworks_summary, plot_renderer=mock_render)
explorer.correlation_plot(
include=["Height (cm)", "Width (cm)", "Depth (cm)"])
explorer.correlation_plot(include=["Height (cm)", "Width (cm)"])
def test_correlation_plot_annotations(artworks_df, artworks_summary):
plt.cla()
def mock_render(fig):
assert len(fig['data']) == 1
corr = [item for row in fig['data'][0]['z'] for item in row]
labels = [l['text'] for l in fig['layout']['annotations']]
for c, l in zip(corr, labels):
assert '{:.2g}'.format(c) == l
explorer = Explorer(artworks_summary, plot_renderer=mock_render)
explorer.correlation_plot(include=['Height (cm)', 'Width (cm)',
'Depth (cm)'])
explorer.correlation_plot(include=['Height (cm)', 'Width (cm)'])
def test_cdf_plot(artworks_df, artworks_summary):
column = "Height (cm)"
plt.cla()
def mock_render(fig):
ax = fig.axes[0]
assert len(ax.lines) == 1
line = ax.lines[0]
tdigest = artworks_summary.tdigest(column)
xs = [tdigest.percentile(p) for p in [0, 100]]
assert line.get_xdata()[0] == xs[0]
assert line.get_xdata()[-1] == xs[-1]
assert line.get_ydata()[0] == 0
assert line.get_ydata()[-1] == 100
explorer = Explorer(artworks_summary, plot_renderer=mock_render)
explorer.cdf_plot("Height (cm)")