def hist(dataset,
xgroup_attr=None,
ygroup_attr=None,
xlim=None,
ylim=None,
noticks=False,
**kwargs):
"""Compute and draw feature histograms (for groups of samples)
This is a convenience wrapper around matplotlib's hist() function. It
supports it entire API, but data is taken from an input dataset. In
addition, feature histograms for groups of dataset samples can be drawn as
an array of subplots. Using ``xgroup_attr`` and ``ygroup_attr`` up to two
sample attributes can be selected and samples groups are defined by their
unique values. For example, plotting histograms for all combinations of
``targets`` and ``chunks`` attribute values in a dataset is done by this
code:
>>> from mvpa2.viz import hist
>>> from mvpa2.misc.data_generators import normal_feature_dataset
>>> ds = normal_feature_dataset(10, 3, 10, 5)
>>> plots = hist(ds, ygroup_attr='targets', xgroup_attr='chunks',
... noticks=None, xlim=(-.5,.5), normed=True)
>>> len(plots)
15
This function can also be used with plain arrays, in which case it will
fall back on the behavior of matplotlib's hist() and additional
functionality is not available.
Parameters
----------
dataset : Dataset or array
xgroup_attr : string, optional
Name of a samples attribute to be used as targets
ygroup_attr : None or string, optional
If a string, a histogram will be plotted per each target and each
chunk (as defined in sa named `chunks_attr`), resulting is a
histogram grid (targets x chunks).
xlim : None or 2-tuple, optional
Common x-axis limits for all histograms.
ylim : None or 2-tuple, optional
Common y-axis limits for all histograms.
noticks : bool or None, optional
If True, no axis ticks will be plotted. If False, each histogram subplot
will have its own ticks. If None, only the outer subplots will
have ticks. This is useful to save space in large plots, but should be
combined with ``xlim`` and ``ylim`` arguments in order to ensure equal
axes across subplots.
**kwargs
Any additional arguments are passed to matplotlib's hist().
Returns
-------
list
List of figure handlers for all generated subplots.
"""
xgroup = {'attr': xgroup_attr}
ygroup = {'attr': ygroup_attr}
for grp in (xgroup, ygroup):
if grp['attr'] is not None and is_datasetlike(dataset):
grp['split'] = ChainNode([
NFoldPartitioner(1, attr=grp['attr']),
Splitter('partitions', attr_values=[2])
])
grp['gen'] = lambda s, x: s.generate(x)
grp['npanels'] = len(dataset.sa[grp['attr']].unique)
else:
grp['split'] = None
grp['gen'] = lambda s, x: [x]
grp['npanels'] = 1
fig = 1
nrows = ygroup['npanels']
ncols = xgroup['npanels']
subplots = []
# for all labels
for row, ds in enumerate(ygroup['gen'](ygroup['split'], dataset)):
for col, d in enumerate(xgroup['gen'](xgroup['split'], ds)):
ax = pl.subplot(nrows, ncols, fig)
if is_datasetlike(d):
data = d.samples
else:
data = d
ax.hist(data.ravel(), **kwargs)
if xlim is not None:
pl.xlim(xlim)
if noticks is True or (noticks is None and row < nrows - 1):
pl.xticks([])
if noticks is True or (noticks is None and col > 0):
pl.yticks([])
if ncols > 1 and row == 0:
pl.title(str(d.sa[xgroup['attr']].unique[0]))
if nrows > 1 and col == 0:
pl.ylabel(str(d.sa[ygroup['attr']].unique[0]))
fig += 1
subplots.append(ax)
return subplots
def hist(dataset, xgroup_attr=None, ygroup_attr=None,
xlim='same', ylim='same', noticks=False,
**kwargs):
"""Compute and draw feature histograms (for groups of samples)
This is a convenience wrapper around matplotlib's hist() function. It
supports it entire API, but data is taken from an input dataset. In
addition, feature histograms for groups of dataset samples can be drawn as
an array of subplots. Using ``xgroup_attr`` and ``ygroup_attr`` up to two
sample attributes can be selected and samples groups are defined by their
unique values. For example, plotting histograms for all combinations of
``targets`` and ``chunks`` attribute values in a dataset is done by this
code:
>>> from mvpa2.viz import hist
>>> from mvpa2.misc.data_generators import normal_feature_dataset
>>> ds = normal_feature_dataset(10, 3, 10, 5)
>>> plots = hist(ds, ygroup_attr='targets', xgroup_attr='chunks',
... noticks=None, xlim=(-.5,.5), normed=True)
>>> len(plots)
15
This function can also be used with plain arrays, in which case it will
fall back on the behavior of matplotlib's hist() and additional
functionality is not available.
Parameters
----------
dataset : Dataset or array
xgroup_attr : string, optional
Name of a samples attribute to be used as targets
ygroup_attr : None or string, optional
If a string, a histogram will be plotted per each target and each
chunk (as defined in sa named `chunks_attr`), resulting is a
histogram grid (targets x chunks).
xlim : None or 2-tuple or 'same', optional
Common x-axis limits for all histograms. By default all plots will have
the same range of values. Set to None if you would like to let them vary.
ylim : None or 2-tuple or 'same', optional
Common y-axis limits for all histograms. If same, heights for all
histograms will be made equal depending on the data.
noticks : bool or None, optional
If True, no axis ticks will be plotted. If False, each histogram subplot
will have its own ticks. If None, only the outer subplots will
have ticks. This is useful to save space in large plots, but should be
combined with ``xlim`` and ``ylim`` arguments in order to ensure equal
axes across subplots.
**kwargs
Any additional arguments are passed to matplotlib's hist().
Returns
-------
list
List of figure handlers for all generated subplots.
"""
externals.exists("pylab", raise_=True)
import pylab as pl
xgroup = {'attr': xgroup_attr}
ygroup = {'attr': ygroup_attr}
for grp in (xgroup, ygroup):
if grp['attr'] is not None and is_datasetlike(dataset):
grp['split'] = ChainNode([NFoldPartitioner(1, attr=grp['attr']),
Splitter('partitions', attr_values=[2])])
grp['gen'] = lambda s, x: s.generate(x)
grp['npanels'] = len(dataset.sa[grp['attr']].unique)
else:
grp['split'] = None
grp['gen'] = lambda s, x: [x]
grp['npanels'] = 1
fig = 1
nrows = ygroup['npanels']
ncols = xgroup['npanels']
subplots = []
ylim_ = (0, 0)
# for all labels
for row, ds in enumerate(ygroup['gen'](ygroup['split'], dataset)):
for col, d in enumerate(xgroup['gen'](xgroup['split'], ds)):
ax = pl.subplot(nrows, ncols, fig)
if is_datasetlike(d):
data = d.samples
else:
data = d
(n, bins, patches) = ax.hist(data.ravel(), **kwargs)
if xlim is not None:
pl.xlim(_get_lim(data, xlim))
if ylim is not None:
if isinstance(ylim, basestring) and ylim.lower() == 'same':
ylim_ = (min(ylim_[0], min(n)), max(ylim_[1], max(n)))
else:
pl.ylim(ylim)
# for 'same' we would have needed to estimate histograms first
# and then adjust their ylim
if noticks is True or (noticks is None and row < nrows - 1):
pl.xticks([])
if noticks is True or (noticks is None and col > 0):
pl.yticks([])
if ncols > 1 and row == 0:
pl.title(str(d.sa[xgroup['attr']].unique[0]))
if nrows > 1 and col == 0:
pl.ylabel(str(d.sa[ygroup['attr']].unique[0]))
fig += 1
subplots.append(ax)
# if we have changed ylim_
if ylim_ != (0, 0):
for ax in subplots:
ax.set_ylim(ylim_)
return subplots
def matshow(matrix,
xlabel_attr=None,
ylabel_attr=None,
numbers=None,
**kwargs):
"""Enhanced version of matplotlib's matshow().
This version is able to handle datasets, and label axis according to
dataset attribute values.
>>> from mvpa2.viz import matshow
>>> from mvpa2.misc.data_generators import normal_feature_dataset
>>> ds = normal_feature_dataset(10, 2, 18, 5)
>>> im = matshow(ds, ylabel_attr='targets', xlabel_attr='chunks',
... numbers='%.0f')
Parameters
----------
matrix : 2D array
The matrix that is to be plotted as an image. If 'matrix' is of
type Dataset the function tries to plot the corresponding samples.
xlabel_attr : str or None
If not 'None' matrix is treated as a Dataset and labels are
extracted from the sample attribute named 'xlabel_attr'.
The labels are used as the 'x_tick_lables' of the image.
ylabel_attr : str or None
If not 'None' matrix is treated as a Dataset and labels are
extracted from the feature attribute named 'ylabel_attr'.
The labels are used as the 'y_tick_lables' of the image.
numbers : dict, str or None
If not 'None' plots matrix values as text inside the image.
If a string is provided, then this string is used as format string.
In case that a dictionary is provided, the dictionary gets passed
on to the text command, and, '%d' is used to format the values.
**kwargs
Additional parameters passed on to matshow().
Returns
-------
matplotlib.AxesImage
Handler for the created image.
"""
externals.exists("pylab", raise_=True)
import pylab as pl
if numbers is not None:
if isinstance(numbers, str):
numbers_format = numbers
numbers_alpha = None
numbers_kwargs_ = {}
elif isinstance(numbers, dict):
numbers_format = '%d'
numbers_alpha = numbers.pop('numbers_alpha', None)
numbers_kwargs_ = numbers
else:
raise TypeError("The argument to keyword 'numbers' must be "
"either of type string or type dict")
_xlabel = None
_ylabel = None
# check if dataset 'is' a confusion matrix
if is_datasetlike(matrix):
if xlabel_attr is not None and ylabel_attr is not None:
_xlabel = matrix.get_attr(xlabel_attr)[0].value # LookupError
_ylabel = matrix.get_attr(
ylabel_attr)[0].value # if it's not there
matrix = np.asanyarray(matrix)
fig = pl.gcf()
ax = pl.gca()
im = ax.matshow(matrix, **kwargs)
# customize labels if _xlabel and _ylabel are set
if _xlabel is not None and _ylabel is not None:
xlabels = [item.get_text() for item in ax.get_xticklabels()]
xlabels[1:-1] = _xlabel
ax.set_xticklabels(xlabels)
pl.xlabel(xlabel_attr)
ylabels = [item.get_text() for item in ax.get_yticklabels()]
ylabels[1:-1] = _ylabel
ax.set_yticklabels(ylabels)
pl.ylabel(ylabel_attr)
# colorbar customization for discrete matrix
# code taken from old ConfusionMatrix.plot()
# TODO: colorbar should be discrete as well
cb_kwargs_ = {}
maxv = np.max(matrix)
if ('int' in matrix.dtype.name) and (maxv > 0):
boundaries = np.linspace(0, maxv, np.min((maxv, 10)), True)
cb_kwargs_['format'] = '%d'
cb_kwargs_['ticks'] = boundaries
cb = pl.colorbar(im, **cb_kwargs_)
# plot matrix values inside the image if number is set
if numbers is not None:
colors = [im.to_rgba(0), im.to_rgba(maxv)]
for i, cas in enumerate(matrix):
for j, v in enumerate(cas):
numbers_kwargs_['color'] = colors[int(v < maxv / 2)]
# code taken from old ConfusionMatrix.plot()
if numbers_alpha is None:
alpha = 1.0
else:
# scale non-linearly w.r.t. value
alpha = 1 - np.array(1 - np.float(v) / maxv)**numbers_alpha
pl.text(j,
i,
numbers_format % v,
alpha=alpha,
**numbers_kwargs_)
return im
def matshow(matrix, xlabel_attr=None, ylabel_attr=None, numbers=None,
**kwargs):
"""Enhanced version of matplotlib's matshow().
This version is able to handle datasets, and label axis according to
dataset attribute values.
>>> from mvpa2.viz import matshow
>>> from mvpa2.misc.data_generators import normal_feature_dataset
>>> ds = normal_feature_dataset(10, 2, 18, 5)
>>> im = matshow(ds, ylabel_attr='targets', xlabel_attr='chunks',
... numbers='%.0f')
Parameters
----------
matrix : 2D array
The matrix that is to be plotted as an image. If 'matrix' is of
type Dataset the function tries to plot the corresponding samples.
xlabel_attr : str or None
If not 'None' matrix is treated as a Dataset and labels are
extracted from the sample attribute named 'xlabel_attr'.
The labels are used as the 'x_tick_lables' of the image.
ylabel_attr : str or None
If not 'None' matrix is treated as a Dataset and labels are
extracted from the feature attribute named 'ylabel_attr'.
The labels are used as the 'y_tick_lables' of the image.
numbers : dict, str or None
If not 'None' plots matrix values as text inside the image.
If a string is provided, then this string is used as format string.
In case that a dictionary is provided, the dictionary gets passed
on to the text command, and, '%d' is used to format the values.
**kwargs
Additional parameters passed on to matshow().
Returns
-------
matplotlib.AxesImage
Handler for the created image.
"""
externals.exists("pylab", raise_=True)
import pylab as pl
if numbers is not None:
if isinstance(numbers, str):
numbers_format = numbers
numbers_alpha = None
numbers_kwargs_ = {}
elif isinstance(numbers, dict):
numbers_format = '%d'
numbers_alpha = numbers.pop('numbers_alpha', None)
numbers_kwargs_ = numbers
else:
raise TypeError("The argument to keyword 'numbers' must be "
"either of type string or type dict")
_xlabel = None
_ylabel = None
# check if dataset 'is' a confusion matrix
if is_datasetlike(matrix):
if xlabel_attr is not None and ylabel_attr is not None:
_xlabel = matrix.get_attr(xlabel_attr)[0].value # LookupError
_ylabel = matrix.get_attr(ylabel_attr)[0].value # if it's not there
matrix = np.asanyarray(matrix)
fig = pl.gcf()
ax = pl.gca()
im = ax.matshow(matrix, **kwargs)
# customize labels if _xlabel and _ylabel are set
if _xlabel is not None and _ylabel is not None:
xlabels = [item.get_text() for item in ax.get_xticklabels()]
xlabels[1:-1] = _xlabel
ax.set_xticklabels(xlabels)
pl.xlabel(xlabel_attr)
ylabels = [item.get_text() for item in ax.get_yticklabels()]
ylabels[1:-1] = _ylabel
ax.set_yticklabels(ylabels)
pl.ylabel(ylabel_attr)
# colorbar customization for discrete matrix
# code taken from old ConfusionMatrix.plot()
# TODO: colorbar should be discrete as well
cb_kwargs_ = {}
maxv = np.max(matrix)
if ('int' in matrix.dtype.name) and (maxv > 0):
boundaries = np.linspace(0, maxv, np.min((maxv, 10)), True)
cb_kwargs_['format'] = '%d'
cb_kwargs_['ticks'] = boundaries
cb = pl.colorbar(im, **cb_kwargs_)
# plot matrix values inside the image if number is set
if numbers is not None:
colors = [im.to_rgba(0), im.to_rgba(maxv)]
for i, cas in enumerate(matrix):
for j, v in enumerate(cas):
numbers_kwargs_['color'] = colors[int(v<maxv/2)]
# code taken from old ConfusionMatrix.plot()
if numbers_alpha is None:
alpha = 1.0
else:
# scale non-linearly w.r.t. value
alpha = 1 - np.array(1 - np.float(v)/maxv) ** numbers_alpha
pl.text(j, i, numbers_format % v,
alpha=alpha, **numbers_kwargs_)
return im
