def test_parallel_coordinates():
data = pd.DataFrame({'x': range(10),
'y': range(10),
'z': range(10),
'c': list('ABABABABAB')})
plot = pdvega.parallel_coordinates(data, 'c', alpha=0.5)
utils.validate_vegalite(plot.spec)
utils.check_encodings(plot.spec, x='variable', y='value',
color='c', detail='index', opacity=utils.IGNORE)
enc = plot.spec['encoding']
assert plot.spec['mark'] == 'line'
assert enc['x']['type'] == 'nominal'
assert enc['y']['type'] == 'quantitative'
assert enc['color']['type'] == 'nominal'
assert enc['detail']['type'] == 'quantitative'
assert enc['opacity']['value'] == 0.5
df = utils.get_data(plot.spec)
assert set(pd.unique(df['variable'])) == {'x', 'y', 'z'}
plot = pdvega.parallel_coordinates(data, 'c', cols=['x', 'y'])
utils.validate_vegalite(plot.spec)
utils.check_encodings(plot.spec, x='variable', y='value',
color='c', detail='index')
enc = plot.spec['encoding']
assert plot.spec['mark'] == 'line'
assert enc['x']['type'] == 'nominal'
assert enc['y']['type'] == 'quantitative'
assert enc['color']['type'] == 'nominal'
assert enc['detail']['type'] == 'quantitative'
df = utils.get_data(plot.spec)
assert set(pd.unique(df['variable'])) == {'x', 'y'}
def test_lag_plot(lag):
data = pd.DataFrame({'x': range(10), 'y': range(10)})
# test series input
plot = pdvega.lag_plot(data['x'], lag=lag)
lag_data = utils.get_data(plot)
spec = plot.to_dict()
assert spec['mark'] == 'point'
assert spec['encoding']['x']['type'] == 'quantitative'
assert spec['encoding']['y']['type'] == 'quantitative'
utils.check_encodings(plot, x='y(t)', y='y(t + {0})'.format(lag))
assert lag_data.shape == (data.shape[0] - lag, 2)
# test dataframe input
plot = pdvega.lag_plot(data, lag=lag)
lag_data = utils.get_data(plot)
spec = plot.to_dict()
assert spec['mark'] == 'point'
assert spec['encoding']['x']['type'] == 'quantitative'
assert spec['encoding']['y']['type'] == 'quantitative'
assert spec['encoding']['color']['type'] == 'nominal'
utils.check_encodings(plot,
x='y(t)',
y='y(t + {0})'.format(lag),
color='variable')
assert lag_data.shape == (2 * (data.shape[0] - lag), 3)
def test_scatter_common_columns():
df = pd.DataFrame({"x": [1, 4, 2, 3, 5], "y": [6, 3, 4, 5, 2]})
plot = df.vgplot.scatter(x="x", y="y", c="y")
utils.validate_vegalite(plot)
assert plot.mark == "point"
utils.check_encodings(plot, x="x", y="y", color="y")
def test_df_hist(stacked, histtype):
df = pd.DataFrame({'x': range(10), 'y': range(10)})
marks = {
'bar': 'bar',
'step': {
'type': 'line',
'interpolate': 'step'
},
'stepfilled': {
'type': 'area',
'interpolate': 'step'
}
}
# bar histogram
plot = df.vgplot.hist(bins=5, stacked=stacked, histtype=histtype)
assert plot.spec['mark'] == marks[histtype]
if stacked:
# No default opacity for a stacked histogram
utils.check_encodings(plot.spec,
x='value',
y=utils.IGNORE,
color='variable')
else:
utils.check_encodings(plot.spec,
x='value',
y=utils.IGNORE,
color='variable',
opacity=utils.IGNORE)
assert plot.spec['encoding']['x']['bin'] == {'maxbins': 5}
assert plot.spec['encoding']['y']['aggregate'] == 'count'
assert plot.spec['encoding']['y']['stack'] == ('zero' if stacked else None)
def test_df_hist(stacked, histtype, maxbins):
df = pd.DataFrame({"x": range(10), "y": range(10)})
marks = {
"bar": "bar",
"step": {
"type": "line",
"interpolate": "step"
},
"stepfilled": {
"type": "area",
"interpolate": "step"
},
}
# bar histogram
plot = df.vgplot.hist(bins=maxbins, stacked=stacked, histtype=histtype)
assert plot.mark == marks[histtype]
if stacked:
# No default opacity for a stacked histogram
utils.check_encodings(plot,
x="value",
y=utils.IGNORE,
color="variable")
else:
utils.check_encodings(plot,
x="value",
y=utils.IGNORE,
color="variable",
opacity=utils.IGNORE)
assert plot["encoding"]["x"]["bin"] == {"maxbins": maxbins}
assert plot["encoding"]["y"]["aggregate"] == "count"
assert plot["encoding"]["y"]["stack"] == ("zero" if stacked else None)
def test_df_kde_y():
df = pd.DataFrame({"x": range(10), "y": range(10)})
plot = df.vgplot.kde(y="y", bw_method="scott")
assert plot.mark == "line"
utils.check_encodings(plot, x=" ", y="Density", color=utils.IGNORE)
data = plot.data
assert set(pd.unique(data["variable"])) == {"y"}
def test_scatter_common_columns():
df = pd.DataFrame({'x': [1, 4, 2, 3, 5], 'y': [6, 3, 4, 5, 2]})
plot = df.vgplot.scatter(x='x', y='y', c='y')
utils.validate_vegalite(plot.spec)
assert plot.spec['mark'] == 'circle'
utils.check_encodings(plot.spec, x='x', y='y', color='y')
def test_andrews_curves():
data = pd.DataFrame({
'x': range(10),
'y': range(10),
'z': range(10),
'c': list('ABABABABAB')
})
n_samples = 120
n_points = len(data)
plot = pdvega.andrews_curves(data, 'c', samples=120, alpha=0.5)
utils.validate_vegalite(plot)
utils.check_encodings(plot,
x='t',
y=' ',
color='c',
detail='sample',
opacity=utils.IGNORE)
spec = plot.to_dict()
enc = spec['encoding']
assert spec['mark'] == 'line'
assert enc['x']['type'] == 'quantitative'
assert enc['y']['type'] == 'quantitative'
assert enc['color']['type'] == 'nominal'
assert enc['detail']['type'] == 'quantitative'
assert enc['opacity']['value'] == 0.5
df = utils.get_data(plot)
assert len(df) == n_samples * n_points
def test_df_kde_y():
df = pd.DataFrame({'x': range(10), 'y': range(10)})
plot = df.vgplot.kde(y='y', bw_method='scott')
assert plot.spec['mark'] == 'line'
utils.check_encodings(plot.spec, x=' ', y='Density', color=utils.IGNORE)
data = utils.get_data(plot.spec)
assert set(pd.unique(data['variable'])) == {'y'}
def test_line_simple():
df = pd.DataFrame({'x': [1, 4, 2, 3, 5], 'y': [6, 3, 4, 5, 2]})
plot = df.vgplot.line()
utils.validate_vegalite(plot.spec)
assert plot.spec['mark'] == 'line'
utils.check_encodings(plot.spec, x='index', y='value', color='variable')
data = utils.get_data(plot.spec)
assert set(pd.unique(data['variable'])) == {'x', 'y'}
def test_df_hexbin_C():
df = pd.DataFrame({'x': range(10), 'y': range(10), 'C': range(10)})
gridsize = 10
plot = df.vgplot.hexbin(x='x', y='y', C='C', gridsize=gridsize)
assert plot.spec['mark'] == 'rect'
utils.check_encodings(plot.spec, x='x', y='y', color='C')
assert plot.spec['encoding']['x']['bin'] == {"maxbins": gridsize}
assert plot.spec['encoding']['y']['bin'] == {"maxbins": gridsize}
assert plot.spec['encoding']['color']['aggregate'] == "mean"
def test_ser_kde():
ser = pd.Series(range(10), name="x")
plot = ser.vgplot.kde(bw_method="scott")
assert plot.mark == "line"
utils.check_encodings(
plot,
x=' ',
y='x',
)
def test_df_hexbin_C():
df = pd.DataFrame({"x": range(10), "y": range(10), "C": range(10)})
gridsize = 10
plot = df.vgplot.hexbin(x="x", y="y", C="C", gridsize=gridsize)
assert plot.mark == "rect"
utils.check_encodings(plot, x="x", y="y", color="C")
assert plot["encoding"]["x"]["bin"] == alt.Bin(maxbins=gridsize)
assert plot["encoding"]["y"]["bin"] == alt.Bin(maxbins=gridsize)
assert plot["encoding"]["color"]["aggregate"] == "mean"
def test_bar_stacked():
df = pd.DataFrame({"x": [1, 4, 2, 3, 5], "y": [6, 3, 4, 5, 2]})
plot = df.vgplot.bar(stacked=True)
utils.validate_vegalite(plot)
assert plot.mark == "bar"
utils.check_encodings(plot, x="index", y="value", color="variable")
data = plot.data
assert set(pd.unique(data["variable"])) == {"x", "y"}
assert plot["encoding"]["y"]["stack"] == "zero"
def test_bar_stacked():
df = pd.DataFrame({'x': [1, 4, 2, 3, 5], 'y': [6, 3, 4, 5, 2]})
plot = df.vgplot.bar(stacked=True)
utils.validate_vegalite(plot.spec)
assert plot.spec['mark'] == 'bar'
utils.check_encodings(plot.spec, x='index', y='value', color='variable')
data = utils.get_data(plot.spec)
assert set(pd.unique(data['variable'])) == {'x', 'y'}
assert plot.spec['encoding']['y']['stack'] == "zero"
def test_barh_xy():
df = pd.DataFrame({"x": [1, 4, 2, 3, 5], "y": [6, 3, 4, 5, 2]})
plot = df.vgplot.barh(x="x", y="y")
utils.validate_vegalite(plot)
assert plot.mark == "bar"
utils.check_encodings(plot, x="value", y="x", color="variable")
data = plot.data
assert set(pd.unique(data["variable"])) == {"y"}
assert plot["encoding"]["x"]["stack"] is None
def test_line_simple():
df = pd.DataFrame({"x": [1, 4, 2, 3, 5], "y": [6, 3, 4, 5, 2]})
plot = df.vgplot.line()
utils.validate_vegalite(plot)
assert plot.mark == "line"
utils.check_encodings(plot, x="index", y="value", color="variable")
data = plot.data
assert set(pd.unique(data["variable"])) == {"x", "y"}
def test_scatter_color_size():
df = pd.DataFrame({
"x": [1, 4, 2, 3, 5],
"y": [6, 3, 4, 5, 2],
"c": range(5),
"s": range(5)
})
plot = df.vgplot.scatter(x="x", y="y", c="c", s="s")
utils.validate_vegalite(plot)
assert plot.mark == "point"
utils.check_encodings(plot, x="x", y="y", color="c", size="s")
def test_scatter_color_size():
df = pd.DataFrame({
'x': [1, 4, 2, 3, 5],
'y': [6, 3, 4, 5, 2],
'c': range(5),
's': range(5)
})
plot = df.vgplot.scatter(x='x', y='y', c='c', s='s')
utils.validate_vegalite(plot.spec)
assert plot.spec['mark'] == 'circle'
utils.check_encodings(plot.spec, x='x', y='y', color='c', size='s')
def test_df_area_xy_unstacked():
df = pd.DataFrame({
"x": [1, 4, 2, 3, 5],
"y": [6, 3, 4, 5, 2],
"z": range(5)
})
plot = df.vgplot.area(x="x", y="y", stacked=False)
utils.validate_vegalite(plot)
assert plot.mark == "area"
utils.check_encodings(plot, x="x", y="value", color="variable")
data = plot.data
assert set(pd.unique(data["variable"])) == {"y"}
assert plot["encoding"]["y"]["stack"] is None
def test_barh_simple():
df = pd.DataFrame({'x': [1, 4, 2, 3, 5], 'y': [6, 3, 4, 5, 2]})
plot = df.vgplot.barh()
utils.validate_vegalite(plot.spec)
assert plot.spec['mark'] == 'bar'
utils.check_encodings(plot.spec,
y='index',
x='value',
color='variable',
opacity=utils.IGNORE)
data = utils.get_data(plot.spec)
assert set(pd.unique(data['variable'])) == {'x', 'y'}
assert plot.spec['encoding']['x']['stack'] is None
def test_df_area_xy_unstacked():
df = pd.DataFrame({
'x': [1, 4, 2, 3, 5],
'y': [6, 3, 4, 5, 2],
'z': range(5)
})
plot = df.vgplot.area(x='x', y='y', stacked=False)
utils.validate_vegalite(plot.spec)
assert plot.spec['mark'] == 'area'
utils.check_encodings(plot.spec, x='x', y='value', color='variable')
data = utils.get_data(plot.spec)
assert set(pd.unique(data['variable'])) == {'y'}
assert plot.spec['encoding']['y']['stack'] is None
def test_df_area_unstacked():
df = pd.DataFrame({"x": [1, 4, 2, 3, 5], "y": [6, 3, 4, 5, 2]})
plot = df.vgplot.area(stacked=False)
utils.validate_vegalite(plot)
assert plot.mark == "area"
utils.check_encodings(plot,
x="index",
y="value",
color="variable",
opacity=utils.IGNORE)
data = plot.data
assert set(pd.unique(data["variable"])) == {"x", "y"}
assert plot["encoding"]["y"]["stack"] is None
assert plot["encoding"]["opacity"]["value"] == 0.7
def test_df_area_unstacked():
df = pd.DataFrame({'x': [1, 4, 2, 3, 5], 'y': [6, 3, 4, 5, 2]})
plot = df.vgplot.area(stacked=False)
utils.validate_vegalite(plot.spec)
assert plot.spec['mark'] == 'area'
utils.check_encodings(plot.spec,
x='index',
y='value',
color='variable',
opacity=utils.IGNORE)
data = utils.get_data(plot.spec)
assert set(pd.unique(data['variable'])) == {'x', 'y'}
assert plot.spec['encoding']['y']['stack'] is None
assert plot.spec['encoding']['opacity']['value'] == 0.7
def test_series_hist(histtype):
ser = pd.Series(range(10))
marks = {
'bar': 'bar',
'step': {
'type': 'line',
'interpolate': 'step'
},
'stepfilled': {
'type': 'area',
'interpolate': 'step'
}
}
plot = ser.vgplot.hist(bins=5, histtype=histtype)
assert plot.spec['mark'] == marks[histtype]
utils.check_encodings(plot.spec, x='0', y=utils.IGNORE)
assert plot.spec['encoding']['x']['bin'] == {'maxbins': 5}
assert plot.spec['encoding']['y']['aggregate'] == 'count'
def test_series_hist(histtype, maxbins):
ser = pd.Series(range(10))
marks = {
"bar": "bar",
"step": {
"type": "line",
"interpolate": "step"
},
"stepfilled": {
"type": "area",
"interpolate": "step"
},
}
plot = ser.vgplot.hist(bins=maxbins, histtype=histtype)
assert plot.mark == marks[histtype]
utils.check_encodings(plot, x="0", y=utils.IGNORE)
assert plot["encoding"]["x"]["bin"] == {"maxbins": maxbins}
assert plot["encoding"]["y"]["aggregate"] == "count"
def test_series_barh():
ser = pd.Series([4, 5, 4, 5], index=['A', 'B', 'C', 'D'])
plot = ser.vgplot.barh()
utils.validate_vegalite(plot.spec)
assert plot.spec['mark'] == 'bar'
utils.check_encodings(plot.spec, y='index', x='0')
def test_df_hexbin_Cfunc():
df = pd.DataFrame({'x': range(10), 'y': range(10), 'C': range(10)})
plot = df.vgplot.hexbin(x='x', y='y', C='C', reduce_C_function=min)
utils.check_encodings(plot.spec, x='x', y='y', color='C')
assert plot.spec['encoding']['color']['aggregate'] == "min"
def test_series_line():
ser = pd.Series([3, 2, 3, 2, 3])
plot = ser.vgplot.line()
utils.validate_vegalite(plot.spec)
assert plot.spec['mark'] == 'line'
utils.check_encodings(plot.spec, x='index', y='0')
def test_ser_kde():
ser = pd.Series(range(10), name='x')
plot = ser.vgplot.kde(bw_method='scott')
assert plot.spec['mark'] == 'line'
utils.check_encodings(plot.spec, x=' ', y='x')
