def kde(data, bins=None, kernel=None, **kwargs):
""" kde(a, bins=None, range=None, **kwargs)
Make a kernerl density estimate plot of the data. This is like a
histogram, but produces a smoother result, thereby better represening
the probability density function.
See the vv.StatData for more statistics on data.
Parameters
----------
a : array_like
The data to calculate the historgam of.
bins : int (optional)
The number of bins. If not given, the best number of bins is
determined automatically using the Freedman-Diaconis rule.
kernel : float or sequence (optional)
The kernel to use for distributing the values. If a scalar is given,
a Gaussian kernel with a sigma equal to the given number is used.
If not given, the best kernel is chosen baded on the number of bins.
kwargs : keyword arguments
These are given to the plot function.
"""
# Get stats
from visvis.processing.statistics import StatData
stats = StatData(data)
# Get kde
xx, values = stats.kde(bins, kernel)
# Plot
return vv.plot(xx, values, **kwargs)
def kde(data, bins=None, kernel=None, **kwargs):
""" kde(a, bins=None, range=None, **kwargs)
Make a kernerl density estimate plot of the data. This is like a
histogram, but produces a smoother result, thereby better represening
the probability density function.
See the vv.StatData for more statistics on data.
Parameters
----------
a : array_like
The data to calculate the historgam of.
bins : int (optional)
The number of bins. If not given, the best number of bins is
determined automatically using the Freedman-Diaconis rule.
kernel : float or sequence (optional)
The kernel to use for distributing the values. If a scalar is given,
a Gaussian kernel with a sigma equal to the given number is used.
If not given, the best kernel is chosen baded on the number of bins.
kwargs : keyword arguments
These are given to the plot function.
"""
# Get stats
from visvis.processing.statistics import StatData
stats = StatData(data)
# Get kde
xx, values = stats.kde(bins, kernel)
# Plot
return vv.plot(xx, values, **kwargs)
def hist(data, bins=None, drange=None, normed=False, weights=None):
""" hist(a, bins=None, range=None, normed=False, weights=None)
Make a histogram plot of the data. Uses np.histogram (new version)
internally. See its docs for more information.
See the kde() function for a more accurate density estimate.
See the vv.StatData for more statistics on data.
Parameters
----------
a : array_like
The data to calculate the historgam of.
bins : int or sequence of scalars, optional
If `bins` is an int, it defines the number of equal-width bins in
the given range. If `bins` is a sequence, it defines the bin edges,
including the rightmost edge, allowing for non-uniform bin widths.
If bins is not given, the best number of bins is determined
automatically using the Freedman-Diaconis rule.
range : (float, float)
The lower and upper range of the bins. If not provided, range is
simply (a.min(), a.max()). Values outside the range are ignored.
normed : bool
If False, the result will contain the number of samples in each bin.
If True, the result is the value of the probability *density*
function at the bin, normalized such that the *integral* over the
range is 1. Note that the sum of the histogram values will not be
equal to 1 unless bins of unity width are chosen; it is not a
probability *mass* function.
weights : array_like
An array of weights, of the same shape as `a`. Each value in `a`
only contributes its associated weight towards the bin count
(instead of 1). If `normed` is True, the weights are normalized,
so that the integral of the density over the range remains 1.
"""
# Auto determine bins?
if bins is None:
from visvis.processing.statistics import StatData
stats = StatData(data)
bins = stats.best_number_of_bins()
# let numpy do the work
if np.__version__ < '1.3':
values, edges = np.histogram(data, bins, drange, normed, weights, new=True)
else:
values, edges = np.histogram(data, bins, drange, normed, weights)
# the bins are the left bin edges, let's get the centers
centers = np.empty(values.shape, np.float64)
for i in range(len(values)):
centers[i] = (edges[i] + edges[i+1]) * 0.5
# plot
dbin = centers[1] - centers[0]
return vv.bar(centers, values, width=dbin*0.9)
def hist(data, bins=None, drange=None, normed=False, weights=None):
""" hist(a, bins=None, range=None, normed=False, weights=None)
Make a histogram plot of the data. Uses np.histogram (new version)
internally. See its docs for more information.
See the kde() function for a more accurate density estimate.
See the vv.StatData for more statistics on data.
Parameters
----------
a : array_like
The data to calculate the historgam of.
bins : int or sequence of scalars, optional
If `bins` is an int, it defines the number of equal-width bins in
the given range. If `bins` is a sequence, it defines the bin edges,
including the rightmost edge, allowing for non-uniform bin widths.
If bins is not given, the best number of bins is determined
automatically using the Freedman-Diaconis rule.
range : (float, float)
The lower and upper range of the bins. If not provided, range is
simply (a.min(), a.max()). Values outside the range are ignored.
normed : bool
If False, the result will contain the number of samples in each bin.
If True, the result is the value of the probability *density*
function at the bin, normalized such that the *integral* over the
range is 1. Note that the sum of the histogram values will not be
equal to 1 unless bins of unity width are chosen; it is not a
probability *mass* function.
weights : array_like
An array of weights, of the same shape as `a`. Each value in `a`
only contributes its associated weight towards the bin count
(instead of 1). If `normed` is True, the weights are normalized,
so that the integral of the density over the range remains 1.
"""
# Auto determine bins?
if bins is None:
from visvis.processing.statistics import StatData
stats = StatData(data)
bins = stats.best_number_of_bins()
# let numpy do the work
values, edges = np.histogram(data, bins, drange, normed, weights)
# the bins are the left bin edges, let's get the centers
centers = np.empty(values.shape, np.float64)
for i in range(len(values)):
centers[i] = (edges[i] + edges[i + 1]) * 0.5
# plot
dbin = centers[1] - centers[0]
return vv.bar(centers, values, width=dbin * 0.9)
def __init__(self, data, width, whiskers):
# Get stats of data
self._stats = StatData(data)
# Init width
self._width = width
# Init whisker style
self._whiskers = whiskers
# Init line style
self._lc = (0, 0, 0)
self._lw = 1
# Calculate now
self.calculate()
box.SetWhiskers(value)
return locals()
if __name__ == '__main__':
vv.figure(1)
vv.clf()
a = vv.gca()
d1 = np.random.normal(1, 4, (1000, 1000))
d2 = np.random.normal(2, 3, (20, ))
d3 = np.random.uniform(-1, 3, (100, ))
d4 = [
1, 2, 1, 2.0, 8, 2, 3, 1, 2, 2, 3, 2, 2.1, 8, 8, 8, 8, 8, 1.2, 1.3, 0,
0, 1.5, 2
]
b = boxplot((d1, d2, d3, d4), width=0.8, whiskers='violin')
##
dd = d4
stat = StatData(dd)
bins1, values1 = stat.histogram_np(normed=True)
bins2, values2 = stat.histogram()
bins3, values3 = stat.kde()
vv.figure(2)
vv.clf()
vv.bar(bins2, values2) #, lc='r', ms='.', mc='r')
vv.plot(bins3, values3) #, lc='g', ms='.', mc='g')
vv.plot(bins1, values1, lc='b', ms='.', mc='b', ls=':', mw=4)
#print abs(bins1-bins2).sum()
""" Get/Set the style of the whiskers.
"""
def fget(self):
return self._boxes[0]._whiskers
def fset(self, value):
for box in self._boxes:
box.SetWhiskers(value)
return locals()
if __name__ == '__main__':
vv.figure(1); vv.clf()
a = vv.gca()
d1 = np.random.normal(1, 4, (1000,1000))
d2 = np.random.normal(2, 3, (20,))
d3 = np.random.uniform(-1, 3, (100,))
d4 = [1,2,1,2.0, 8, 2, 3, 1, 2, 2, 3, 2, 2.1, 8, 8, 8, 8, 8, 1.2, 1.3, 0, 0, 1.5, 2]
b = boxplot((d1,d2,d3, d4), width=0.8, whiskers='violin')
##
dd = d4
stat = StatData(dd)
bins1, values1 = stat.histogram_np(normed=True)
bins2, values2 = stat.histogram()
bins3, values3 = stat.kde( )
vv.figure(2); vv.clf()
vv.bar(bins2, values2)#, lc='r', ms='.', mc='r')
vv.plot(bins3, values3)#, lc='g', ms='.', mc='g')
vv.plot(bins1, values1, lc='b', ms='.', mc='b', ls=':', mw=4)
#print abs(bins1-bins2).sum()
