def test_dodge_preserve_single_text():
df1 = pd.DataFrame({'x': ['a', 'b', 'b', 'b'], 'y': ['a', 'a', 'b', 'b']})
d = position_dodge(preserve='single', width=0.9)
p = (ggplot(df1, aes('x', fill='y')) + geom_bar(position=d) +
geom_text(aes(y=after_stat('count'), label=after_stat('count')),
stat='count',
position=d,
va='bottom'))
assert p + _theme == 'dodge_preserve_single_text'
def test_points():
p = (
p0 +
geom_point(aes(fill=after_stat('density'), size=after_stat('density')),
stat='density_2d',
stroke=0,
n=16,
contour=False) + scale_size_radius(range=(0, 6)))
assert p == 'points'
def test_calculated_aes():
# after_stat('ae')
mapping1 = aes('x', y=after_stat('density'))
mapping2 = aes('x', y=after_stat('density*2'))
mapping3 = aes('x', y=after_stat('density + count'))
mapping4 = aes('x', y=after_stat('func(density)'))
def _test():
assert list(mapping1._calculated.keys()) == ['y']
assert list(mapping2._calculated.keys()) == ['y']
assert list(mapping3._calculated.keys()) == ['y']
assert list(mapping4._calculated.keys()) == ['y']
assert mapping1['y'].after_stat == 'density'
assert mapping2['y'].after_stat == 'density*2'
assert mapping3['y'].after_stat == 'density + count'
assert mapping4['y'].after_stat == 'func(density)'
assert mapping1._calculated['y'] == 'density'
assert mapping2._calculated['y'] == 'density*2'
assert mapping3._calculated['y'] == 'density + count'
assert mapping4._calculated['y'] == 'func(density)'
_test()
# 'stat(ae)', DEPRECATED but still works
mapping1 = aes('x', y='stat(density)')
mapping2 = aes('x', y='stat(density*2)')
mapping3 = aes('x', y='stat(density + count)')
mapping4 = aes('x', y='stat(func(density))')
_test()
# '..ae..', DEPRECATED but still works
mapping1 = aes('x', y='..density..')
mapping2 = aes('x', y='..density..*2')
mapping3 = aes('x', y='..density.. + ..count..')
mapping4 = aes('x', y='func(..density..)')
_test()
df = pd.DataFrame({'x': [1, 2, 2, 3, 3, 3, 4, 4, 4, 4]})
p = ggplot(df) + geom_bar(aes(x='x', fill=after_stat('count + 2')))
p.draw_test()
p = ggplot(df) + geom_bar(aes(x='x', fill='stat(count + 2)'))
p.draw_test()
p = ggplot(df) + geom_bar(aes(x='x', fill='..count.. + 2'))
p.draw_test()
def test_stat_count_float():
df = pd.DataFrame({'x': ['a', 'b'], 'weight': [1.5, 2.5]})
p = (ggplot(df) + aes(x='x', weight='weight', fill='x') + geom_bar() +
geom_text(aes(label=after_stat('count')), stat='count'))
assert p + _theme == 'stat-count-float'
def test_computed_y_column():
p = (ggplot(df, aes('x'))
+ stat_ecdf(size=2)
# Should be able to used computed y column & create a
# new mapped column also called y
+ stat_ecdf(aes(y=after_stat('y-0.2')), size=2, color='blue')
)
assert p == 'computed_y_column'
from scipy.stats import norm
import plotnine as plt
#----------------------------------------
# Generate data
#----------------------------------------
data = pd.DataFrame(data=np.random.normal(-1, 1, 4000))
data = data[data<0].dropna()
data = -1*data[0:2000].reset_index(drop=True)
data.columns = ['v1']
#----------------------------------------
# Figure 1
#----------------------------------------
data['pdf'] = norm.pdf(data, 1, 1)/norm.sf(0, 1, 1)
hist_plot = plt.ggplot(data, plt.aes(x='v1', y=plt.after_stat('density')))+ plt.geom_histogram(alpha=0.6, fill='red') + plt.theme_bw() + plt.geom_line(plt.aes(x='v1', y='pdf'))
model2 = lpdensity(data['v1'], bw=[0.5], grid = np.linspace(0, 4, num=int(4/0.5)))
model2.plot()
#-------------------------------------------------------------------
# lpdensity(): Estimation with bandwidth 0.5 on provided grid points
#-------------------------------------------------------------------
model1 = lpdensity(data['v1'], bw=[0.5], grid = np.linspace(0, 4, num=int(4/0.5)))
model1.plot()
print(repr(model1))
#----------------------------------------
# lpdensity(): extracting estimation
# results
#----------------------------------------
def test_polygon():
p = p0 + stat_density_2d(aes(fill=after_stat('level')), geom='polygon')
assert p == 'polygon'
def test_contours():
p = p0 + geom_density_2d(aes(color=after_stat('level')))
assert p == 'contours'
