This file is indexed.

/usr/share/pyshared/mvpa2/clfs/gda.py is in python-mvpa2 2.1.0-1.

This file is owned by root:root, with mode 0o644.

The actual contents of the file can be viewed below.

  1
  2
  3
  4
  5
  6
  7
  8
  9
 10
 11
 12
 13
 14
 15
 16
 17
 18
 19
 20
 21
 22
 23
 24
 25
 26
 27
 28
 29
 30
 31
 32
 33
 34
 35
 36
 37
 38
 39
 40
 41
 42
 43
 44
 45
 46
 47
 48
 49
 50
 51
 52
 53
 54
 55
 56
 57
 58
 59
 60
 61
 62
 63
 64
 65
 66
 67
 68
 69
 70
 71
 72
 73
 74
 75
 76
 77
 78
 79
 80
 81
 82
 83
 84
 85
 86
 87
 88
 89
 90
 91
 92
 93
 94
 95
 96
 97
 98
 99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
# emacs: -*- mode: python; py-indent-offset: 4; indent-tabs-mode: nil -*-
# vi: set ft=python sts=4 ts=4 sw=4 et:
### ### ### ### ### ### ### ### ### ### ### ### ### ### ### ### ### ### ### ##
#
#   See COPYING file distributed along with the PyMVPA package for the
#   copyright and license terms.
#
### ### ### ### ### ### ### ### ### ### ### ### ### ### ### ### ### ### ### ##
"""Gaussian Discriminant Analyses: LDA and QDA

   Basic implementation at the moment: no data sphering, nor
   dimensionality reduction tricks are in place ATM
"""

"""
TODO:

 * too much in common with GNB -- LDA/QDA/GNB could reuse much of machinery
 * provide actual probabilities computation as in GNB
 * LDA/QDA -- make use of data sphering and may be operating in the
              subspace of centroids

Was based on GNB code
"""

__docformat__ = 'restructuredtext'

import numpy as np

from numpy import ones, zeros, sum, abs, isfinite, dot
from mvpa2.base import warning, externals
from mvpa2.clfs.base import Classifier, accepts_dataset_as_samples
from mvpa2.base.learner import DegenerateInputError
from mvpa2.base.param import Parameter
from mvpa2.base.state import ConditionalAttribute
#from mvpa2.measures.base import Sensitivity


if __debug__:
    from mvpa2.base import debug

__all__ = [ "LDA", "QDA" ]

class GDA(Classifier):
    """Gaussian Discriminant Analysis -- base for LDA and QDA

    """

    __tags__ = ['binary', 'multiclass']


    prior = Parameter('laplacian_smoothing',
             allowedtype='basestring',
             choices=["laplacian_smoothing", "uniform", "ratio"],
             doc="""How to compute prior distribution.""")


    def __init__(self, **kwargs):
        """Initialize a GDA classifier.
        """

        # init base class first
        Classifier.__init__(self, **kwargs)

        # pylint friendly initializations
        self.means = None
        """Means of features per class"""
        self.cov = None
        """Co-variances per class, but "vars" is taken ;)"""
        self.ulabels = None
        """Labels classifier was trained on"""
        self.priors = None
        """Class probabilities"""
        self.nsamples_per_class = None
        """Number of samples per class - used by derived classes"""

        # Define internal state of classifier
        self._norm_weight = None

    def _get_priors(self, nlabels, nsamples, nsamples_per_class):
        """Return prior probabilities given data
        """
        prior = self.params.prior
        if prior == 'uniform':
            priors = np.ones((nlabels,))/nlabels
        elif prior == 'laplacian_smoothing':
            priors = (1+np.squeeze(nsamples_per_class)) \
                          / (float(nsamples) + nlabels)
        elif prior == 'ratio':
            priors = np.squeeze(nsamples_per_class) / float(nsamples)
        else:
            raise ValueError, \
                  "No idea on how to handle '%s' way to compute priors" \
                  % self.params.prior
        return priors


    def _train(self, dataset):
        """Train the classifier using `dataset` (`Dataset`).
        """
        params = self.params
        targets_sa_name = self.get_space()
        targets_sa = dataset.sa[targets_sa_name]

        # get the dataset information into easy vars
        X = dataset.samples
        labels = targets_sa.value
        self.ulabels = ulabels = targets_sa.unique
        nlabels = len(ulabels)
        label2index = dict((l, il) for il, l in enumerate(ulabels))

        # set the feature dimensions
        nsamples = len(X)
        nfeatures = dataset.nfeatures

        self.means = means = \
                     np.zeros((nlabels, nfeatures))
        # degenerate dimension are added for easy broadcasting later on
        # XXX might want to remove -- for now taken from GNB as is
        self.nsamples_per_class = nsamples_per_class \
                                  = np.zeros((nlabels, 1))
        self.cov = cov = \
                     np.zeros((nlabels, nfeatures, nfeatures))


        # Estimate cov
        # better loop than repmat! ;)
        for l, il in label2index.iteritems():
            Xl = X[labels == l]
            nsamples_per_class[il] = len(Xl)
            # TODO: degenerate case... no samples for known label for
            #       some reason?
            means[il] = np.mean(Xl, axis=0)
            # since we have means already lets do manually cov here
            Xldm = Xl - means[il]
            cov[il] = np.dot(Xldm.T, Xldm)
            # scaling will be done correspondingly in LDA or QDA

        # Store prior probabilities
        self.priors = self._get_priors(nlabels, nsamples, nsamples_per_class)

        if __debug__ and 'GDA' in debug.active:
            debug('GDA', "training finished on data.shape=%s " % (X.shape, )
                  + "min:max(data)=%f:%f" % (np.min(X), np.max(X)))


    def _untrain(self):
        """Untrain classifier and reset all learnt params
        """
        self.means = None
        self.cov = None
        self.ulabels = None
        self.priors = None
        super(GDA, self)._untrain()


    @accepts_dataset_as_samples
    def _predict(self, data):
        """Predict the output for the provided data.
        """
        params = self.params

        self.ca.estimates = prob_cs_cp = self._g_k(data)

        # Take the class with maximal (log)probability
        # XXX in GNB it is axis=0, i.e. classes were first
        winners = prob_cs_cp.argmax(axis=1)
        predictions = [self.ulabels[c] for c in winners]

        if __debug__ and 'GDA' in debug.active:
            debug('GDA', "predict on data.shape=%s min:max(data)=%f:%f " %
                  (data.shape, np.min(data), np.max(data)))

        return predictions


class LDA(GDA):
    """Linear Discriminant Analysis.
    """

    __tags__ = GDA.__tags__ + ['linear', 'lda']


    def _untrain(self):
        self._w = None
        self._b = None
        super(LDA, self)._untrain()


    def _train(self, dataset):
        super(LDA, self)._train(dataset)
        nlabels = len(self.ulabels)
        # Sum and scale the covariance
        self.cov = cov = \
            np.sum(self.cov, axis=0) \
            / (np.sum(self.nsamples_per_class) - nlabels)

        # For now as simple as that -- see notes on top
        try:
            covi = np.linalg.inv(cov)
        except Exception, e:
            raise DegenerateInputError, \
                  "Data is probably singular, since inverse fails. Got %s"\
                  % (e,)

        # Precompute and store the actual separating hyperplane and offset
        self._w = np.dot(covi, self.means.T)
        self._b = b = np.zeros((nlabels,))
        for il in xrange(nlabels):
            m = self.means[il]
            b[il] = np.log(self.priors[il]) - 0.5 * np.dot(np.dot(m.T, covi), m)

    def _g_k(self, data):
        """Return decision function values"""
        return np.dot(data, self._w) + self._b


class QDA(GDA):
    """Quadratic Discriminant Analysis.
    """

    __tags__ = GDA.__tags__ + ['non-linear', 'qda']

    def _untrain(self):
        # XXX theoretically we could use the same _w although with
        # different "content"
        self._icov = None
        self._b = None
        super(QDA, self)._untrain()

    def _train(self, dataset):
        super(QDA, self)._train(dataset)

        # XXX should we drag cov around at all then?
        self._icov = np.zeros(self.cov.shape)

        for ic, cov in enumerate(self.cov):
            cov /= float(self.nsamples_per_class[ic])
            try:
                self._icov[ic] = np.linalg.inv(cov)
            except Exception, e:
                raise DegenerateInputError, \
                      "Data is probably singular, since inverse fails. Got %s"\
                      % (e,)

        self._b = np.array([np.log(p) - 0.5 * np.log(np.linalg.det(c))
                            for p,c in zip(self.priors, self.cov)])

    def _g_k(self, data):
        """Return decision function values"""
        res = []
        for m, covi, b in zip(self.means, self._icov, self._b):
            dm = data - m
            res.append(b - 0.5 * np.sum(np.dot(dm, covi) * dm, axis=1))
        return np.array(res).T