def __init__(self, which_set, center=False, multi_target=False):
"""
:param which_set: one of ['train','test']
:param center: data is in range [0,256], center=True subtracts 127.5.
:param multi_target: load extra information as additional labels.
"""
assert which_set in ['train', 'test']
X = FoveatedNORB.load(which_set)
# put things in pylearn2's DenseDesignMatrix format
X = numpy.cast['float32'](X)
#this is uint8
y = NORBSmall.load(which_set, 'cat')
if multi_target:
y_extra = NORBSmall.load(which_set, 'info')
y = numpy.hstack((y[:, numpy.newaxis], y_extra))
if center:
X -= 127.5
view_converter = retina.RetinaCodingViewConverter((96, 96, 2),
(8, 4, 2, 2))
super(FoveatedNORB, self).__init__(X=X,
y=y,
view_converter=view_converter)
def __init__(self,
which_set,
center=False,
scale=False,
start=None,
stop=None,
one_hot=False,
restrict_instances=None,
preprocessor=None):
"""
:param which_set: one of ['train','test']
:param center: data is in range [0,256], center=True subtracts 127.5.
:param multi_target: load extra information as additional labels.
"""
if which_set not in ['train', 'test']:
raise ValueError("Unrecognized which_set value: " + which_set)
X = FoveatedNORB.load(which_set)
# put things in pylearn2's DenseDesignMatrix format
X = numpy.cast['float32'](X)
#this is uint8
y = NORBSmall.load(which_set, 'cat')
y_extra = NORBSmall.load(which_set, 'info')
assert y_extra.shape[0] == y.shape[0]
instance = y_extra[:, 0]
assert instance.min() >= 0
assert instance.max() <= 9
self.instance = instance
if center:
X -= 127.5
if scale:
X /= 127.5
else:
if scale:
X /= 255.
view_converter = retina.RetinaCodingViewConverter((96, 96, 2),
(8, 4, 2, 2))
super(FoveatedNORB, self).__init__(X=X,
y=y,
view_converter=view_converter,
preprocessor=preprocessor)
if one_hot:
self.convert_to_one_hot()
if restrict_instances is not None:
assert start is None
assert stop is None
self.restrict_instances(restrict_instances)
self.restrict(start, stop)
self.y = self.y.astype('float32')
def __init__(self,
which_set,
center=False,
scale=False,
start=None,
stop=None,
one_hot=False,
restrict_instances=None,
preprocessor=None):
self.args = locals()
if which_set not in ['train', 'test']:
raise ValueError("Unrecognized which_set value: " + which_set)
X = FoveatedNORB.load(which_set)
X = numpy.cast['float32'](X)
# this is uint8
y = NORBSmall.load(which_set, 'cat')
y_extra = NORBSmall.load(which_set, 'info')
assert y_extra.shape[0] == y.shape[0]
instance = y_extra[:, 0]
assert instance.min() >= 0
assert instance.max() <= 9
self.instance = instance
if center:
X -= 127.5
if scale:
X /= 127.5
else:
if scale:
X /= 255.
view_converter = retina.RetinaCodingViewConverter((96, 96, 2),
(8, 4, 2, 2))
super(FoveatedNORB, self).__init__(X=X,
y=y,
view_converter=view_converter,
preprocessor=preprocessor)
if one_hot:
self.convert_to_one_hot()
if restrict_instances is not None:
assert start is None
assert stop is None
self.restrict_instances(restrict_instances)
self.restrict(start, stop)
self.y = self.y.astype('float32')