def test_save_all_figures():
data = dav.DataVisualization()
builder = dav.DesignerDataVisualization(df_nan)
data.builder = builder
data.visual_missing_data()
builder.figure.save_all_figures(folder="tmp")
assert len(list(Path("tmp").glob("*.png"))) == 4
def plotdata(
df: pd.DataFrame,
plot_type: str = "lineplot",
dark_mode: bool = False,
grid: bool = False,
figsize: tuple = (12, 8),
dpi: int = 300,
save_fig: bool = True,
show_fig: bool = False,
x_label: str = None,
y_label: str = None,
hue: str = None,
palette: Union[str, list, dict] = None,
folder: str = None,
**kwargs,
) -> None:
"""Plotting all pandas dataframes as `oneliner`.
!!! warning
All defined functions of`ai2business.visualization.data_visualization` will be
called via `getattr`, so be sure that the `class DataVisualization` function
is typed correctly.
Args:
df (pd.DataFrame): [description]
plot_type (str, optional): [description]. Defaults to "lineplot".
dark_mode (bool, optional): [description]. Defaults to False.
grid (bool, optional): [description]. Defaults to False.
figsize (tuple, optional): [description]. Defaults to (12, 8).
dpi (int, optional): [description]. Defaults to 300.
save_fig (bool, optional): [description]. Defaults to True.
show_fig (bool, optional): [description]. Defaults to False.
x_label (str, optional): [description]. Defaults to None.
y_label (str, optional): [description]. Defaults to None.
hue (str, optional): [description]. Defaults to None.
palette (Union[str, list, dict], optional): [description]. Defaults to None.
folder (str, optional): [description]. Defaults to None.
"""
data = data_visualization.DataVisualization()
builder = data_visualization.DesignerDataVisualization(
df,
dark_mode=dark_mode,
grid=grid,
figsize=figsize,
dpi=dpi,
x_label=x_label,
y_label=y_label,
hue=hue,
palette=palette,
)
data.builder = builder
try:
getattr(data, plot_type.lower())(**kwargs)
except AttributeError as exc:
print(f"ERROR: {exc}")
if save_fig:
builder.figure.save_all_figures(folder=folder)
if show_fig:
plt.show()
def test_lineplot_whitegrid():
data = dav.DataVisualization()
builder = dav.DesignerDataVisualization(iris_dataframe, grid=True)
data.builder = builder
data.lineplot()
folder = "tmp_whitegrid"
builder.figure.save_all_figures(folder=folder)
assert len(list(Path(f"{folder}").glob("*.png"))) == 1
def test_violinplot():
data = dav.DataVisualization()
builder = dav.DesignerDataVisualization(
df_dict_fruits["get_interest_over_time"], )
data.builder = builder
data.violinplot()
folder = f"{test_violinplot.__name__}"
builder.figure.save_all_figures(folder=folder)
assert len(list(Path(f"{folder}").glob("*.png"))) == 1
def test_visual_missing_data():
data = dav.DataVisualization()
builder = dav.DesignerDataVisualization(df_nan)
data.builder = builder
data.visual_missing_data()
result = builder.figure.return_product
for i, key in enumerate(result.keys()):
result[key].savefig(Path(f"./test_visual_missing_data_{key}.png"))
assert len(list(Path(".").glob("test_visual_missing_data_*.png"))) == i + 1
def test_histogramplot():
data = dav.DataVisualization()
builder = dav.DesignerDataVisualization(
df_dict_fruits["get_interest_over_time"],
x_label="apple",
hue="super market")
data.builder = builder
data.histogramplot(multiple="stack", log_scale=True)
folder = f"{test_histogramplot.__name__}"
builder.figure.save_all_figures(folder=folder)
assert len(list(Path(f"{folder}").glob("*.png"))) == 1
def test_pairmap_complex():
data = dav.DataVisualization()
builder = dav.DesignerDataVisualization(
df_dict_fruits["get_interest_over_time"].drop(columns="isPartial"),
x_label="apple",
y_label="pineapple",
)
data.builder = builder
data.pairmap(complex=True)
folder = f"{test_pairmap_complex.__name__}"
builder.figure.save_all_figures(folder=folder)
assert len(list(Path(f"{folder}").glob("*.png"))) == 1
def test_correlation_2():
data = dav.DataVisualization()
builder = dav.DesignerDataVisualization(
df_dict_fruits["get_interest_over_time"],
x_label="apple",
y_label="pineapple",
)
data.builder = builder
data.correlationmap(diagonal=True)
folder = f"{test_correlation_2.__name__}"
builder.figure.save_all_figures(folder=folder)
assert len(list(Path(f"{folder}").glob("*.png"))) == 1
def test_list_product_parts():
# Test return of products
data = dav.DataVisualization()
builder = dav.DesignerDataVisualization(df_nan)
data.builder = builder
data.visual_missing_data()
result = builder.figure.list_product_parts
assert (
result
== "Product parts: get_nullity_matrix, get_nullity_bar, get_nullity_heatmap, get_nullity_dendrogram"
)
def test_residualplot():
data = dav.DataVisualization()
builder = dav.DesignerDataVisualization(
df_dict_fruits["get_interest_over_time"],
x_label="apple",
y_label="super market",
)
data.builder = builder
data.residualplot(lowess=True, color="b")
folder = f"{test_residualplot.__name__}"
builder.figure.save_all_figures(folder=folder)
assert len(list(Path(f"{folder}").glob("*.png"))) == 1
def test_hexagonplot():
data = dav.DataVisualization()
builder = dav.DesignerDataVisualization(
df_dict_fruits["get_interest_over_time"],
x_label="apple",
y_label="pineapple",
)
data.builder = builder
data.hexagonplot(color="#4CB391")
folder = f"{test_hexagonplot.__name__}"
builder.figure.save_all_figures(folder=folder)
assert len(list(Path(f"{folder}").glob("*.png"))) == 1
def test_categoryplot():
data = dav.DataVisualization()
builder = dav.DesignerDataVisualization(
df_dict_fruits["get_interest_over_time"],
palette="dark",
x_label="apple",
y_label="pineapple",
)
data.builder = builder
data.categoryplot()
folder = f"{test_categoryplot.__name__}"
builder.figure.save_all_figures(folder=folder)
assert len(list(Path(f"{folder}").glob("*.png"))) == 1
def test_distributionplot():
# Test distributionplot with new data set:
data = dav.DataVisualization()
builder = dav.DesignerDataVisualization(
df_dict_fruits["get_related_queries"]["apple"]["top"],
palette="dark",
x_label="value",
)
data.builder = builder
data.distributionplot(kind="ecdf")
folder = f"{test_distributionplot.__name__}"
builder.figure.save_all_figures(folder=folder)
assert len(list(Path(f"{folder}").glob("*.png"))) == 1
def test_pointplot():
# Test pointplot with new data set:
data = dav.DataVisualization()
builder = dav.DesignerDataVisualization(
df_dict_fruits["get_interest_over_time"],
x_label="apple",
y_label="pineapple",
)
data.builder = builder
data.pointplot()
folder = f"{test_pointplot.__name__}"
builder.figure.save_all_figures(folder=folder)
assert len(list(Path(f"{folder}").glob("*.png"))) == 1
def test_boxenplot():
data = dav.DataVisualization()
builder = dav.DesignerDataVisualization(
df_dict_fruits["get_interest_over_time"],
palette=sns.light_palette("purple"),
dark_mode=True,
grid=True,
)
data.builder = builder
data.boxplot(multiboxen=True)
folder = f"{test_boxenplot.__name__}"
builder.figure.save_all_figures(folder=folder)
assert len(list(Path(f"{folder}").glob("*.png"))) == 1
def test_scatterplot():
data = dav.DataVisualization()
builder = dav.DesignerDataVisualization(
df_dict_fruits["get_interest_over_time"],
x_label="apple",
y_label="pineapple",
hue="super market",
palette="ch:r=-.2,d=.3_r",
)
data.builder = builder
data.scatterplot(size="home delivery")
folder = f"{test_scatterplot.__name__}"
builder.figure.save_all_figures(folder=folder)
assert len(list(Path(f"{folder}").glob("*.png"))) == 1
def test_clustermap():
df = sns.load_dataset("brain_networks", header=[0, 1, 2], index_col=0)
# Select a subset of the networks
used_networks = [1, 5, 6, 7, 8, 12, 13, 17]
used_columns = (
df.columns.get_level_values("network").astype(int).isin(used_networks))
df = df.loc[:, used_columns]
data = dav.DataVisualization()
builder = dav.DesignerDataVisualization(df)
data.builder = builder
data.clustermap()
folder = f"{test_clustermap.__name__}"
builder.figure.save_all_figures(folder=folder)
assert len(list(Path(f"{folder}").glob("*.png"))) == 1
def test_swarmplot():
data = dav.DataVisualization()
builder = dav.DesignerDataVisualization(
df_dict_fruits["get_interest_over_time"],
palette="Set2",
)
data.builder = builder
data.swarmplot(
size=2,
marker="D",
edgecolor="gray",
alpha=0.25,
)
folder = f"{test_swarmplot.__name__}"
builder.figure.save_all_figures(folder=folder)
assert len(list(Path(f"{folder}").glob("*.png"))) == 1
