def processing_dataset_train(self):
current_directory = (os.path.dirname(__file__))
data_directory_path = os.path.join(current_directory, '../../../',
'data')
print("data directory :", data_directory_path)
filePath = os.path.join(data_directory_path, self.test_file)
print("Filepath :", filePath)
train_labels, train_features = process_data(filePath)
print("len of labels {}".format(len(train_labels)))
return train_labels, train_features
def processing_dataset_test(self):
current_directory = (os.path.dirname(__file__))
data_directory_path = os.path.join(current_directory, '../../../',
'data')
print("data directory :", data_directory_path)
filePath = os.path.join(data_directory_path, self.test_file)
print("Filepath :", filePath)
test_filePath = os.path.join(data_directory_path, self.test_file)
test_labels, test_features = process_data(test_filePath)
print("Dataset Processing Done")
return test_labels, test_features
for i in range(len(feature_pos)):
logit += self.top_k.get_value_for_key(feature_pos[i]) * feature_val[i]
a = self.sigmoid(logit)
if a > 0.5:
return 1
else:
return 0
if __name__ == '__main__':
h = hpy()
current_directory = (os.path.dirname(__file__))
data_directory_path = os.path.join(current_directory, '..', 'data')
fileName = "rcv1_train.binary"
filePath = os.path.join(data_directory_path, fileName)
labels, features = process_data(filePath)
D = 47236
lgr = LogisticRegression(num_features=D)
print("len of labels {}".format(len(labels)))
start_time = time.time()
for epoch in range(0, 1):
print("epoch {}".format(epoch))
for i in range(len(labels)):
print("i {}".format(i))
label = labels[i]
label = (1 + label) / 2
example_features = features[i]
feature_pos = [item[0] for item in example_features]
feature_vals = [item[1] for item in example_features]
loss = lgr.train_with_sketch(feature_pos, feature_vals, label)
print("loss {}".format(loss))