def makeQuantiles(series, n):
"""
Compute quantiles of input series.
Parameters
----------
series: Series
Must have 'order' method and index
n: int
Number of quantile buckets
Returns
-------
(edges, quantiles)
edges: ith bucket --> (left edge, right edge)
quantiles: ith bucket --> set of values
"""
series = remove_na(series).copy()
series = series.order()
quantiles = {}
edges = {}
T = float(len(series))
inc = T / n
for i in range(n):
theSlice = series[inc*i:(i+1)*inc]
quantiles[i+1] = theSlice
edges[i+1] = theSlice[0], theSlice[-1]
return edges, quantiles
def makeQuantiles(series, n):
"""
Compute quantiles of input series.
Parameters
----------
series: Series
Must have 'order' method and index
n: int
Number of quantile buckets
Returns
-------
(edges, quantiles)
edges: ith bucket --> (left edge, right edge)
quantiles: ith bucket --> set of values
"""
series = remove_na(series).copy()
series = series.order()
quantiles = {}
edges = {}
T = float(len(series))
inc = T / n
for i in range(n):
theSlice = series[inc * i:(i + 1) * inc]
quantiles[i + 1] = theSlice
edges[i + 1] = theSlice[0], theSlice[-1]
return edges, quantiles
def percentileRank(frame, column=None, kind='mean'):
"""
Return score at percentile for each point in time (cross-section)
Parameters
----------
frame: DataFrame
column: string or Series, optional
Column name or specific Series to compute percentiles for.
If not provided, percentiles are computed for all values at each
point in time. Note that this can take a LONG time.
kind: {'rank', 'weak', 'strict', 'mean'}, optional
This optional parameter specifies the interpretation of the
resulting score:
- "rank": Average percentage ranking of score. In case of
multiple matches, average the percentage rankings of
all matching scores.
- "weak": This kind corresponds to the definition of a cumulative
distribution function. A percentileofscore of 80%
means that 80% of values are less than or equal
to the provided score.
- "strict": Similar to "weak", except that only values that are
strictly less than the given score are counted.
- "mean": The average of the "weak" and "strict" scores, often used in
testing. See
http://en.wikipedia.org/wiki/Percentile_rank
See also
--------
scipy.stats.percentileofscore
Returns
-------
TimeSeries or DataFrame, depending on input
"""
from scipy.stats import percentileofscore
fun = lambda xs, score: percentileofscore(remove_na(xs),
score, kind=kind)
results = {}
framet = frame.T
if column is not None:
if isinstance(column, Series):
for date, xs in frame.T.iteritems():
results[date] = fun(xs, column.get(date, NaN))
else:
for date, xs in frame.T.iteritems():
results[date] = fun(xs, xs[column])
results = Series(results)
else:
for column in frame.columns:
for date, xs in framet.iteritems():
results.setdefault(date, {})[column] = fun(xs, xs[column])
results = DataFrame(results).T
return results
def plot_group(grouped, ax):
keys, values = zip(*grouped)
keys = [_stringify(x) for x in keys]
values = [remove_na(v) for v in values]
ax.boxplot(values, **kwds)
if kwds.get('vert', 1):
ax.set_xticklabels(keys, rotation=rot, fontsize=fontsize)
else:
ax.set_yticklabels(keys, rotation=rot, fontsize=fontsize)
def percentileRank(frame, column=None, kind='mean'):
"""
Return score at percentile for each point in time (cross-section)
Parameters
----------
frame: DataFrame
column: string or Series, optional
Column name or specific Series to compute percentiles for.
If not provided, percentiles are computed for all values at each
point in time. Note that this can take a LONG time.
kind: {'rank', 'weak', 'strict', 'mean'}, optional
This optional parameter specifies the interpretation of the
resulting score:
- "rank": Average percentage ranking of score. In case of
multiple matches, average the percentage rankings of
all matching scores.
- "weak": This kind corresponds to the definition of a cumulative
distribution function. A percentileofscore of 80%
means that 80% of values are less than or equal
to the provided score.
- "strict": Similar to "weak", except that only values that are
strictly less than the given score are counted.
- "mean": The average of the "weak" and "strict" scores, often used in
testing. See
http://en.wikipedia.org/wiki/Percentile_rank
See also
--------
scipy.stats.percentileofscore
Returns
-------
TimeSeries or DataFrame, depending on input
"""
from scipy.stats import percentileofscore
fun = lambda xs, score: percentileofscore(remove_na(xs), score, kind=kind)
results = {}
framet = frame.T
if column is not None:
if isinstance(column, Series):
for date, xs in frame.T.iteritems():
results[date] = fun(xs, column.get(date, NaN))
else:
for date, xs in frame.T.iteritems():
results[date] = fun(xs, xs[column])
results = Series(results)
else:
for column in frame.columns:
for date, xs in framet.iteritems():
results.setdefault(date, {})[column] = fun(xs, xs[column])
results = DataFrame(results).T
return results
def boxplot(self, df, axis=0, secondary_y=False, *args, **kwargs):
if axis == 1:
df = df.T
index = df.columns
self.set_index(index)
clean_values = [remove_na(x) for x in df.values.T]
ax = self.ax
if secondary_y:
ax = self.get_right_ax()
# positions need to start at 0 to align with DateLocator
ax.boxplot(clean_values, positions=np.arange(len(index)))
self.setup_datetime(index)
self.set_formatter()
def boxplot(self, df, axis=0, secondary_y=False, *args, **kwargs):
"""
Currently supports plotting DataFrames.
Downside is that this only works for data that has equal columns.
For something like plotting groups with varying sizes, you'd
need to use boxplot(list()). Example is creating a SeriesGroupBy.boxplot
"""
if axis == 1:
df = df.T
index = df.columns
self.set_index(index)
clean_values = [remove_na(x) for x in df.values.T]
ax = self.find_ax(secondary_y, kwargs)
# positions need to start at 0 to align with TimestampLocator
ax.boxplot(clean_values, positions=np.arange(len(index)))
self.setup_datetime(index)
self.set_formatter()
def boxplot(self, df, axis=0, secondary_y=False, *args, **kwargs):
"""
Currently supports plotting DataFrames.
Downside is that this only works for data that has equal columns.
For something like plotting groups with varying sizes, you'd
need to use boxplot(list()). Example is creating a SeriesGroupBy.boxplot
"""
if axis == 1:
df = df.T
index = df.columns
self.set_index(index)
clean_values = [remove_na(x) for x in df.values.T]
ax = self.find_ax(secondary_y, kwargs)
# positions need to start at 0 to align with TimestampLocator
ax.boxplot(clean_values, positions=np.arange(len(index)))
self.setup_datetime(index)
self.set_formatter()
def skipna_wrapper(x):
nona = remove_na(x)
if len(nona) == 0:
return np.nan
return alternative(nona)
def skipna_wrapper(x):
nona = remove_na(x)
if len(nona) == 0:
return np.nan
return alternative(nona)
def test_remove_na_deprecation(self):
# see gh-16971
with tm.assert_produces_warning(FutureWarning):
remove_na(Series([]))
def test_remove_na_deprecation(self):
# see gh-16971
with tm.assert_produces_warning(FutureWarning):
remove_na(Series([]))
def boxplot(data, column=None, by=None, ax=None, fontsize=None,
rot=0, grid=True, figsize=None, **kwds):
"""
Make a box plot from DataFrame column optionally grouped b ysome columns or
other inputs
Parameters
----------
data : DataFrame or Series
column : column name or list of names, or vector
Can be any valid input to groupby
by : string or sequence
Column in the DataFrame to group by
fontsize : int or string
rot : label rotation angle
kwds : other plotting keyword arguments to be passed to matplotlib boxplot
function
Returns
-------
ax : matplotlib.axes.AxesSubplot
"""
from pandas import Series, DataFrame
if isinstance(data, Series):
data = DataFrame({'x' : data})
column = 'x'
def plot_group(grouped, ax):
keys, values = zip(*grouped)
keys = [_stringify(x) for x in keys]
values = [remove_na(v) for v in values]
ax.boxplot(values, **kwds)
if kwds.get('vert', 1):
ax.set_xticklabels(keys, rotation=rot, fontsize=fontsize)
else:
ax.set_yticklabels(keys, rotation=rot, fontsize=fontsize)
if column == None:
columns = None
else:
if isinstance(column, (list, tuple)):
columns = column
else:
columns = [column]
if by is not None:
if not isinstance(by, (list, tuple)):
by = [by]
fig, axes = _grouped_plot_by_column(plot_group, data, columns=columns,
by=by, grid=grid, figsize=figsize)
# Return axes in multiplot case, maybe revisit later # 985
ret = axes
else:
if ax is None:
ax = _gca()
fig = ax.get_figure()
data = data._get_numeric_data()
if columns:
cols = columns
else:
cols = data.columns
keys = [_stringify(x) for x in cols]
# Return boxplot dict in single plot case
clean_values = [remove_na(x) for x in data[cols].values.T]
bp = ax.boxplot(clean_values, **kwds)
if kwds.get('vert', 1):
ax.set_xticklabels(keys, rotation=rot, fontsize=fontsize)
else:
ax.set_yticklabels(keys, rotation=rot, fontsize=fontsize)
ax.grid(grid)
ret = bp
fig.subplots_adjust(bottom=0.15, top=0.9, left=0.1, right=0.9, wspace=0.2)
return ret
