Ejemplo n.º 1
0
    def _customize_doc(cls):
        #cdoc_old = cls.__doc__
        # Need to append documentation to __init__ method
        idoc_old = cls.__init__.__doc__

        idoc = """
SVM/SVR definition is dependent on specifying kernel, implementation
type, and parameters for each of them which vary depending on the
choices made.

Desired implementation is specified in `svm_impl` argument. Here
is the list if implementations known to this class, along with
specific to them parameters (described below among the rest of
parameters), and what tasks it is capable to deal with
(e.g. regression, binary and/or multiclass classification):

"""
        # XXX Deprecate
        # To not confuse sphinx -- lets avoid Implementations section
        # %s""" % (_rst_section('Implementations'),)


        class NOSClass(object):
            """Helper -- NothingOrSomething ;)
            If list is not empty -- return its entries within string s
            """
            def __init__(self):
                self.seen = []
            def __call__(self, l, s, empty=''):
                if l is None or not len(l):
                    return empty
                else:
                    lsorted = list(l)
                    lsorted.sort()
                    self.seen += lsorted
                    return s % (', '.join(lsorted))
        NOS = NOSClass()

        # Describe implementations
        idoc += ''.join(
            ['\n%s%s : %s' % (_rst_indentstr, k, v[3])
             + NOS(v[1], "\n" + _rst_indentstr + "  Parameters: %s")
             + NOS(v[2], "%s" % _rst(('','\n')[int(len(v[1])>0)], '')
                   + _rst_indentstr + "  Capabilities: %s")
             for k,v in cls._KNOWN_IMPLEMENTATIONS.iteritems()])

        # Describe kernels
        idoc += """

Kernel choice is specified as a kernel instance with kwargument 'kernel`.
Some kernels (e.g. Linear) might allow computation of per feature
sensitivity.

"""
        # XXX Deprecate
        # %s""" % (_rst_section('Kernels'),)

        #idoc += ''.join(
        #    ['\n%s%s' % (_rst_indentstr, k)
        #     + ('', ' : provides sensitivity')[int(v[2] is not None)]
        #     + '\n    ' + NOS(v[1], '%s', 'No parameters')
        #     for k,v in cls._KERNELS.iteritems()])

        # Finally parameters
        NOS.seen += cls._KNOWN_PARAMS# + cls._KNOWN_KERNEL_PARAMS

        idoc += '\n' + _rst_section('Parameters') + '\n' + '\n'.join(
            [v._paramdoc()
             for k,v in cls._SVM_PARAMS.iteritems()
             if k in NOS.seen])

        cls.__dict__['__init__'].__doc__ = handle_docstring(idoc_old) + idoc
Ejemplo n.º 2
0
    def _customize_doc(cls):
        #cdoc_old = cls.__doc__
        # Need to append documentation to __init__ method
        idoc_old = cls.__init__.__doc__

        idoc = """
SVM/SVR definition is dependent on specifying kernel, implementation
type, and parameters for each of them which vary depending on the
choices made.

Desired implementation is specified in `svm_impl` argument. Here
is the list if implementations known to this class, along with
specific to them parameters (described below among the rest of
parameters), and what tasks it is capable to deal with
(e.g. regression, binary and/or multiclass classification):

"""

        # XXX Deprecate
        # To not confuse sphinx -- lets avoid Implementations section
        # %s""" % (_rst_section('Implementations'),)

        class NOSClass(object):
            """Helper -- NothingOrSomething ;)
            If list is not empty -- return its entries within string s
            """
            def __init__(self):
                self.seen = []

            def __call__(self, l, s, empty=''):
                if l is None or not len(l):
                    return empty
                else:
                    lsorted = list(l)
                    lsorted.sort()
                    self.seen += lsorted
                    return s % (', '.join(lsorted))

        NOS = NOSClass()

        # Describe implementations
        idoc += ''.join([
            '\n%s%s : %s' % (_rst_indentstr, k, v[3]) +
            NOS(v[1], "\n" + _rst_indentstr + "  Parameters: %s") + NOS(
                v[2], "%s" % _rst(('', '\n')[int(len(v[1]) > 0)], '') +
                _rst_indentstr + "  Capabilities: %s")
            for k, v in cls._KNOWN_IMPLEMENTATIONS.iteritems()
        ])

        # Describe kernels
        idoc += """

Kernel choice is specified as a kernel instance with kwargument 'kernel`.
Some kernels (e.g. Linear) might allow computation of per feature
sensitivity.

"""
        # XXX Deprecate
        # %s""" % (_rst_section('Kernels'),)

        #idoc += ''.join(
        #    ['\n%s%s' % (_rst_indentstr, k)
        #     + ('', ' : provides sensitivity')[int(v[2] is not None)]
        #     + '\n    ' + NOS(v[1], '%s', 'No parameters')
        #     for k,v in cls._KERNELS.iteritems()])

        # Finally parameters
        NOS.seen += cls._KNOWN_PARAMS  # + cls._KNOWN_KERNEL_PARAMS

        idoc += '\n' + _rst_section('Parameters') + '\n' + '\n'.join([
            v._paramdoc()
            for k, v in cls._SVM_PARAMS.iteritems() if k in NOS.seen
        ])

        cls.__dict__['__init__'].__doc__ = handle_docstring(idoc_old) + idoc