def testKmeans(self):
model = ml.kmeans(self.mtcars, 2, 5, 10)
self.assertIsInstance(model, ml.KMeansModel)
self.assertIsInstance(model.centers, pd.DataFrame)
self.assertEqual(len(model.centers), 2)
self.assertItemsEqual(model.centers.columns.tolist(), self.mtcars.colnames)
self.assertIsInstance(model.predict(range(0, self.mtcars.ncol)), float)
with self.assertRaises(Py4JJavaError):
model.predict([0, 1, 2])
def testKmeans(self):
model = ml.kmeans(self.mtcars, 2, 5, 10)
self.assertIsInstance(model, ml.KMeansModel)
self.assertIsInstance(model.centers, pd.DataFrame)
self.assertEqual(len(model.centers), 2)
self.assertItemsEqual(model.centers.columns.tolist(),
self.mtcars.colnames)
self.assertIsInstance(model.predict(range(0, self.mtcars.ncol)), float)
with self.assertRaises(Py4JJavaError):
model.predict([0, 1, 2])
print('Columns: ' + ', '.join(ddf.colnames))
print('Number of columns: {}'.format(ddf.cols))
print('Number of rows: {}'.format(ddf.rows))
print(ddf.summary())
print(ddf.head(2))
print(ddf.aggregate(['sum(mpg)', 'min(hp)'], ['vs', 'am']))
print(ddf.five_nums())
print(ddf.sample(3))
# Kmeans
km = ml.kmeans(ddf)
clu = km.predict(range(0, ddf.ncol))
print clu
# Linear Regression
lr = ml.linear_regression_gd(ddf)
lr.summary()
# Logistic Regression
lr = ml.logistic_regression_gd(ddf)
lr.summary()
dm.shutdown()