class TestDataset(unittest.TestCase):
def setUp(self):
root_dir = os.path.dirname(os.path.realpath(__file__))
data_dir = os.path.join(root_dir, "data")
self.reader = CsvReader(data_dir)
def test_read_spy(self):
dataset = self.reader.read_dataset(["SPY"], [ColumnName.ADJCLOSE])
self.assertEqual(dataset.column(ColumnName.ADJCLOSE).shape[0], 50)
self.assertEqual(dataset.column(ColumnName.ADJCLOSE).shape[1], 1)
def test_read_goog(self):
dataset = self.reader.read_dataset(["GOOG"], [ColumnName.ADJCLOSE])
self.assertEqual(dataset.column(ColumnName.ADJCLOSE).shape[0], 30)
self.assertEqual(dataset.column(ColumnName.ADJCLOSE).shape[1], 1)
def test_read_gld(self):
dataset = self.reader.read_dataset(["GLD"], [ColumnName.ADJCLOSE])
self.assertEqual(dataset.column(ColumnName.ADJCLOSE).shape[0], 70)
self.assertEqual(dataset.column(ColumnName.ADJCLOSE).shape[1], 1)
def test_read_all(self):
dataset = self.reader.read_dataset(["SPY", "GOOG", "GLD"],
[ColumnName.ADJCLOSE])
self.assertEqual(dataset.column(ColumnName.ADJCLOSE).shape[0], 70)
self.assertEqual(dataset.column(ColumnName.ADJCLOSE).shape[1], 3)
def test_read_multycol(self):
dataset = self.reader.read_dataset(
["SPY"], [ColumnName.CLOSE, ColumnName.ADJCLOSE])
self.assertEqual(len(dataset.columns), 2)
class TestColumn(unittest.TestCase):
def setUp(self):
root_dir = os.path.dirname(os.path.realpath(__file__))
data_dir = os.path.join(root_dir, "data")
self.reader = CsvReader(data_dir)
def test_read_spy(self):
df = self.reader.read_column(["SPY"], ColumnName.ADJCLOSE)
self.assertEqual(df.shape[0], 50)
self.assertEqual(df.shape[1], 1)
def test_read_goog(self):
df = self.reader.read_column(["GOOG"], ColumnName.ADJCLOSE)
self.assertEqual(df.shape[0], 30)
self.assertEqual(df.shape[1], 1)
def test_read_gld(self):
df = self.reader.read_column(["GLD"], ColumnName.ADJCLOSE)
self.assertEqual(df.shape[0], 70)
self.assertEqual(df.shape[1], 1)
def test_read_all(self):
df = self.reader.read_column(["SPY", "GOOG", "GLD"],
ColumnName.ADJCLOSE)
self.assertEqual(df.shape[0], 70)
self.assertEqual(df.shape[1], 3)
def setUp(self):
root_dir = os.path.dirname(os.path.realpath(__file__))
data_dir = os.path.join(root_dir, "data")
self.dfs = [
CsvReader(data_dir).read_column(["SPY"], ColumnName.ADJCLOSE),
CsvReader(data_dir).read_column(["GOOG"], ColumnName.ADJCLOSE)
]
class TestMerge(unittest.TestCase):
def setUp(self):
root_dir = os.path.dirname(os.path.realpath(__file__))
data_dir = os.path.join(root_dir, "data")
self.reader = CsvReader(data_dir)
def test_merge_tickers(self):
df_a = self.reader.read_column(["GOOG"], ColumnName.ADJCLOSE)
df_b = self.reader.read_column(["GLD"], ColumnName.ADJCLOSE)
df_c = Merge().process(df_a, df_b)
self.assertListEqual(df_c.columns.tolist(), ["GOOG", "GLD"])
class TestNormalize(unittest.TestCase):
def setUp(self):
root_dir = os.path.dirname(os.path.realpath(__file__))
data_dir = os.path.join(root_dir, "data")
self.df = CsvReader(data_dir).read_column(["SPY"], ColumnName.ADJCLOSE)
def test_normalize(self):
normalized_df = Normalize().process(self.df)
self.assertEqual(normalized_df.max().tolist(),
(self.df.max() / self.df.iloc[0]).tolist())
def test_indicator(self):
normalized_df = NormalizeIndicator().process(self.df)
self.assertEqual(normalized_df.max().tolist(),
((self.df.max() - self.df.mean()) /
self.df.std()).tolist())
class TestBaseline(unittest.TestCase):
def setUp(self):
root_dir = os.path.dirname(os.path.realpath(__file__))
data_dir = os.path.join(root_dir, "data")
self.df = CsvReader(data_dir).read_column(["SPY", "GOOG", "GLD"],
ColumnName.ADJCLOSE)
def test_fill(self):
self.assertEqual(self.df.isnull().sum().sum(), 60)
filled_df = FillMissing().process(self.df)
self.assertEqual(filled_df.isnull().sum().sum(), 0)
def setUp(self):
root_dir = os.path.dirname(os.path.realpath(__file__))
data_dir = os.path.join(root_dir, "data")
self.reader = CsvReader(data_dir)