In [14]:

    

import gzip
import pickle
import numpy
import timeit
import theano


    

In [15]:

    

def shared_dataset(data_xy, borrow=True):

   data_x, data_y = data_xy
   shared_x = theano.shared(numpy.asarray(data_x,
                                          dtype=theano.config.floatX),
                            borrow=borrow)
   shared_y = theano.shared(numpy.asarray(data_y,
                                          dtype=theano.config.floatX),
                            borrow=borrow)
   return shared_x, theano.tensor.cast(shared_y, 'int32')


    

In [31]:

    

start_time = timeit.default_timer()
DATA_PATH = "/Users/myt007/git/svrg/ni/"
DATA_PATH = ""

train_x_raw = gzip.open(DATA_PATH+"train_x.txt.gz", 'rb')
train_x = pickle.load(train_x_raw)
train_y_raw = gzip.open(DATA_PATH+"train_y.txt.gz", 'rb')
train_y = pickle.load(train_y_raw)
test_x_raw = gzip.open(DATA_PATH+"test_x.txt.gz", 'rb')
test_x = pickle.load(test_x_raw)
test_y_raw = gzip.open(DATA_PATH+"test_y.txt.gz", 'rb')
test_y = pickle.load(test_y_raw)


    

In [29]:

    

print(train_x.shape)
print(train_y.shape)
print(test_x.shape)
print(test_y.shape)


    

    




(494021, 41)
(494021,)
(311029, 41)
(311029,)

In [26]:

    

train_x.dtype


    

    
Out[26]:





dtype('float32')

In [17]:

    

train_set = (train_x, train_y)
valid_set = (train_x, train_y)
test_set = (test_x, test_y)

train_set_x, train_set_y = shared_dataset(train_set)
valid_set_x, valid_set_y = shared_dataset(valid_set)
test_set_x, test_set_y = shared_dataset(test_set)


    

In [18]:

    




    

    
Out[18]:





<TensorType(float64, matrix)>

In [ ]:

    




    

In [ ]:

    




    

In [ ]:

    




    

In [ ]: