def plot(self,
x: Optional[Iterable],
kind: str = 'line',
color: str = 'C0',
mean: bool = False,
median: bool = False,
mode: bool = False,
std: bool = False,
ax: Optional[Axes] = None,
**kwargs) -> Axes:
"""
Plot the function.
:param x: Range of values of x to plot p(x) over.
:param kind: Kind of plot e.g. 'bar', 'line'.
:param color: Optional color for the series.
:param mean: Whether to show marker and label for the mean.
:param median: Whether to show marker and label for the median.
:param mode: Whether to show marker and label for the mode.
:param std: Whether to show marker and label for the standard deviation.
:param ax: Optional matplotlib axes to plot on.
:param kwargs: Additional arguments for the matplotlib plot function.
"""
if x is None:
if (hasattr(self._parent, 'lower_bound')
and hasattr(self._parent, 'upper_bound')):
x = linspace(self._parent.lower_bound,
self._parent.upper_bound, 1001)
else:
raise ValueError('Must pass x if distribution has no bounds.')
data: Series = self.at(x)
axf = AxesFormatter(axes=ax)
ax = axf.axes
if self._method_name in ('pdf', 'cdf', 'logpdf'):
if 'label' not in kwargs.keys():
kwargs['label'] = self._parent.label
data.plot(kind=kind, color=color, ax=axf.axes, **kwargs)
else:
raise ValueError('plot not implemented for {}'.format(self._name))
# stats
y_min = axf.get_y_min()
y_max = axf.get_y_max()
x_mean = self._distribution.mean()
if mean:
axf.add_v_lines(x=x_mean,
y_min=y_min,
y_max=y_max,
line_styles='--',
colors=color)
axf.add_text(x=x_mean,
y=self._distribution.pdf(x_mean),
text=f'mean={x_mean: 0.3f}',
color=color,
h_align='center',
v_align='bottom')
if median:
x_median = self._distribution.median()
axf.add_v_lines(x=x_median,
y_min=y_min,
y_max=y_max,
line_styles='-.',
colors=color)
axf.add_text(x=x_median,
y=self._distribution.pdf(x_median),
text=f'median={x_median: 0.3f}',
color=color,
h_align='center',
v_align='bottom')
if mode:
x_mode = self._parent.mode()
axf.add_v_lines(x=x_mode,
y_min=y_min,
y_max=y_max,
line_styles='-.',
colors=color)
axf.add_text(x=x_mode,
y=self._distribution.pdf(x_mode),
text=f'mode={x_mode: 0.3f}',
color=color,
h_align='center',
v_align='bottom')
if std:
x_std = self._distribution.std()
axf.add_v_lines(x=[x_mean - x_std, x_mean + x_std],
y_min=y_min,
y_max=y_max,
line_styles=':',
colors=color)
axf.add_text(x=x_mean - x_std / 2,
y=self._distribution.pdf(x_mean - x_std / 2),
text=f'std={x_std: 0.3f}',
color=color,
h_align='center',
v_align='bottom')
ax.set_xlabel(self._parent.x_label)
if self._parent.y_label:
ax.set_ylabel(self._parent.y_label)
else:
if self._method_name == 'pdf':
ax.set_ylabel('P(X = x)')
elif self._method_name == 'cdf':
ax.set_ylabel('P(X ≤ x)')
elif self._method_name == 'logpdf':
ax.set_ylabel('log P(X = x)')
else:
ax.set_ylabel(self._name)
return ax
def plot(self,
k: Optional[Iterable[int]],
color: str = 'C0',
kind: str = 'bar',
mean: bool = False,
median: bool = False,
std: bool = False,
ax: Optional[Axes] = None,
**kwargs) -> Axes:
"""
Plot the function.
:param k: Range of values of k to plot p(k) over.
:param color: Optional color for the series.
:param kind: Kind of plot e.g. 'bar', 'line'.
:param mean: Whether to show marker and label for the mean.
:param median: Whether to show marker and label for the median.
:param std: Whether to show marker and label for the standard deviation.
:param ax: Optional matplotlib axes to plot on.
:param kwargs: Additional arguments for the matplotlib plot function.
"""
if k is None:
if (hasattr(self._parent, 'lower_bound')
and hasattr(self._parent, 'upper_bound')):
k = range(self._parent.lower_bound,
self._parent.upper_bound + 1)
else:
raise ValueError('Must pass k if distribution has no bounds.')
data: Series = self.at(k)
axf = AxesFormatter(axes=ax)
ax = axf.axes
# special kwargs
vlines = None
if 'vlines' in kwargs.keys():
vlines = kwargs.pop('vlines')
if 'label' not in kwargs.keys():
kwargs['label'] = self._parent.label
if self._method_name == 'pmf':
data.plot(kind=kind, color=color, ax=axf.axes, **kwargs)
elif self._method_name == 'cdf':
data.plot(kind='line',
color=color,
drawstyle='steps-post',
ax=axf.axes,
**kwargs)
else:
raise ValueError('plot not implemented for {}'.format(self._name))
if vlines:
axf.axes.vlines(x=k, ymin=0, ymax=data.values, color=color)
y_min = axf.get_y_min()
y_max = axf.get_y_max()
x_mean = self._distribution.mean()
if mean:
axf.add_v_lines(x=x_mean,
y_min=y_min,
y_max=y_max,
line_styles='--',
colors=color)
axf.add_text(x=x_mean,
y=self._distribution.pmf(x_mean),
text=f'mean={x_mean: 0.3f}',
color=color,
h_align='center',
v_align='bottom')
if median:
x_median = self._distribution.median()
axf.add_v_lines(x=x_median,
y_min=y_min,
y_max=y_max,
line_styles='-.',
colors=color)
axf.add_text(x=x_median,
y=self._distribution.pmf(x_median),
text=f'median={x_median: 0.3f}',
color=color,
h_align='center',
v_align='bottom')
if std:
x_std = self._distribution.std()
axf.add_v_lines(x=[x_mean - x_std, x_mean + x_std],
y_min=y_min,
y_max=y_max,
line_styles=':',
colors=color)
axf.add_text(x=x_mean - x_std / 2,
y=self._distribution.pmf(x_mean - x_std / 2),
text=f'std={x_std: 0.3f}',
color=color,
h_align='center',
v_align='bottom')
ax.set_xlabel(self._parent.x_label)
if self._parent.y_label:
ax.set_ylabel(self._parent.y_label)
else:
if self._method_name == 'pmf':
ax.set_ylabel('P(K = k)')
elif self._method_name == 'cdf':
ax.set_ylabel('P(K ≤ k)')
else:
ax.set_ylabel(self._name)
return ax