def load_data(self, type='train', batch=0, verbose=2):
"""
Will load the data from the file and will return the data. The important thing to note
is that all the datasets in :mod: ``yann`` all require a ``y`` or a variable to
predict. In case of auto-encoder for instance, the thing to predict is the image
itself. Setup dataset thusly.
Args:
type: ``train``, ``test`` or ``valid``.
default is ``train``
batch: Supply an integer
verbose: Simliar to verbose in toolbox.
Returns:
numpy.ndarray: ``data_x, data_y``
"""
if verbose >= 3:
print "... loading " + type + " data batch " + str(batch)
f = open(self.dataset + '/' + type + '/batch_' + str(batch) + '.pkl',
'rb')
data_x, data_y = cPickle.load(f)
f.close()
if verbose >= 3:
print "... data is loaded"
data_x = check_type(data_x, theano.config.floatX)
data_y = check_type(data_y, theano.config.floatX)
# Theano recommends storing on gpus only as floatX and casts them to ints during use.
# I don't know why, but I am following their recommendations blindly.
return data_x, data_y
def load_data (self, type = 'train', batch = 0, verbose = 2):
"""
Will load the data from the file and will return the data. The important thing to note
is that all the datasets in :mod: ``yann`` all require a ``y`` or a variable to
predict. In case of auto-encoder for instance, the thing to predict is the image
itself. Setup dataset thusly.
Args:
type: ``train``, ``test`` or ``valid``.
default is ``train``
batch: Supply an integer
verbose: Simliar to verbose in toolbox.
Returns:
numpy.ndarray: ``data_x, data_y``
"""
if verbose >= 3:
print "... loading " + type + " data batch " + str(batch)
f = open(self.dataset + '/' + type + '/batch_' + str(batch) +'.pkl', 'rb')
data_x, data_y = cPickle.load(f)
f.close()
if verbose >= 3:
print "... data is loaded"
data_x = check_type (data_x, theano.config.floatX)
data_y = check_type (data_y, theano.config.floatX)
# Theano recommends storing on gpus only as floatX and casts them to ints during use.
# I don't know why, but I am following their recommendations blindly.
return data_x, data_y
def one_hot_labels(self, y, verbose=1):
"""
Function takes in labels and returns a one-hot encoding. Used for max-margin loss.
Args:
y: Labels to be encoded.n_classes
verbose: Typical as in the rest of the toolbox.
Notes:
``self.n_classes``: Number of unique classes in the labels.
This could be found out using the following:
.. code-block: python
import numpy
n_classes = len(numpy.unique(y))
This might be potentially dangerous in case of cached dataset. Although
this is the default if ``n_classes`` is not provided as input to this
module, I discourage anyone from using this.
Returns:
numpy ndarray: one-hot encoded label list.
"""
if self.n_classes is False:
if verbose >= 3:
print "... Making a decision to create n_classes variable, not a good idea."
self.n_classes = len(numpy.unique(y))
# found this technique online somewhere, forgot where couldn't cite.
y1 = -1 * numpy.ones((y.shape[0], self.n_classes))
y1[numpy.arange(y.shape[0]), y] = 1
y1 = check_type(y1, theano.config.floatX)
return y1
def one_hot_labels (self, y, verbose = 1):
"""
Function takes in labels and returns a one-hot encoding. Used for max-margin loss.
Args:
y: Labels to be encoded.n_classes
verbose: Typical as in the rest of the toolbox.
Notes:
``self.n_classes``: Number of unique classes in the labels.
This could be found out using the following:
.. code-block: python
import numpy
n_classes = len(numpy.unique(y))
This might be potentially dangerous in case of cached dataset. Although
this is the default if ``n_classes`` is not provided as input to this
module, I discourage anyone from using this.
Returns:
numpy ndarray: one-hot encoded label list.
"""
if self.n_classes is False:
if verbose >= 3:
print "... Making a decision to create n_classes variable, not a good idea."
self.n_classes = len(numpy.unique(y))
# found this technique online somewhere, forgot where couldn't cite.
y1 = -1 * numpy.ones((y.shape[0], self.n_classes))
y1[numpy.arange(y.shape[0]), y] = 1
y1 = check_type(y1, theano.config.floatX)
return y1
