def test_PutMultipleCol(self):
#Checks to make sure multiple valid columns can be added to a dataframe
with Network() as n:
forex = pd.read_csv("forex.csv")
forex2 = forex.copy()
del forex2['Zero']
del forex2['AskPrice2']
del forex2['BidPrice2']
sourceDict = {'forex': forex, 'forex2': forex2}
df = seriesSource('forex')
newDf = seriesSource('forex2')
columnDictTest = {
'DateTime': forex['DateTime'],
'AskPrice1': forex['AskPrice1'],
'BidPrice1': forex['BidPrice1'],
'AskPrice2': forex['AskPrice2'],
'BidPrice2': forex['BidPrice2'],
'Zero': forex['Zero']
}
columnDict = ['AskPrice2', 'BidPrice2', 'Zero']
putColumns(df, columnDict, newDf, name='test')
toComp = pd.DataFrame(columns=list(forex2))
for key in columnDictTest.keys():
toComp[key] = columnDictTest[key]
sinkDict = piEval(n, sourceDict)
toComp2 = sinkDict['test']
self.assertIsNone(pd.testing.assert_frame_equal(toComp, toComp2))
n.report()
def testLog(self):
with Network() as n:
sourceDict = {'fake': randomWalkSeries()} # here series are loaded
log(seriesSource('fake'))
sinkDict = piEval(n, sourceDict)
n.report()
self.assertTrue(y == np.log(x) for x, y in zip(sourceDict['fake'], sinkDict['__0__']))
def testCeiling2(self):
with Network() as n:
sourceDict = {'fake': randomWalkSeries3()} # here series are loaded
ceiling(seriesSource('fake'))
sinkDict = piEval(n, sourceDict)
n.report()
self.assertTrue(y == ceil(x) for x, y in zip(sourceDict['fake'].values, sinkDict['__0__'].values))
def test_greaterOrEqual_2(self):
with Network() as n:
sourceDict = {'fake': randomWalkSeries()} # here series are loaded
greaterOrEqual(seriesSource('fake'), seriesSource('fake'))
sinkDict = piEval(n, sourceDict)
for x in sinkDict['__0__']:
self.assertTrue(x)
print("End of GE Test 2")
def test_notEqual_1(self):
with Network() as n:
sourceDict = {'fake': randomWalkSeries()} # here series are loaded
notEqual(seriesSource('fake'), seriesSource('fake'))
sinkDict = piEval(n, sourceDict)
for x in sinkDict['__0__']:
self.assertFalse(x)
print("End of NE Test 1")
def testABS(self):
with Network() as n:
sourceDict = {'fake': randomWalkSeries2()} # here series are loaded
abs(seriesSource('fake'))
sinkDict = piEval(n, sourceDict)
n.report()
for x in sinkDict['__0__']:
print(x)
self.assertGreaterEqual(x, 0)
def test_TimeWeightMeanTINY_6Min(self):
with Network() as n:
forex = pd.read_csv('forex-tiny-ez.csv')
sourceDict = {'forex': forex}
df = seriesSource('forex')
timeW = 360
timeWeightMean(df, timewindow=timeW, name="test")
sinkDict = piEval(n, sourceDict)
n.report()
def test_TimeWeightSTDINT(self):
with Network() as n:
forex = pd.read_csv('forex.csv')
sourceDict = {'forex': forex}
df = seriesSource('forex')
timeW = 2
timeWeightSTD(df, timewindow=timeW, name="test")
sinkDict = piEval(n, sourceDict)
n.report()
def test_TimeWeightMeanDateTimeMinutes(self):
with Network() as n:
forex = pd.read_csv('forex.csv')
sourceDict = {'forex': forex}
df = seriesSource('forex')
timeW = timedelta(minutes=5)
timeWeightMean(df, timewindow=timeW, name="test")
sinkDict = piEval(n, sourceDict)
n.report()
def test_STDEV_Series(self):
with Network() as n:
sourceDict = {'fake': randomWalkSeries()} # here series are loaded
df = seriesSource('fake')
stdev(df, name="test")
toComp = sourceDict['fake'].std()
sinkDict = piEval(n, sourceDict)
toComp2 = sinkDict['test']
self.assertEqual(toComp, toComp2)
n.report()
def test_lessorEqual_5(self):
with Network() as n:
sourceDict = {
'fake': randomWalkSeries3()
} # here series are loaded
lessOrEqual(seriesSource('fake'), seriesSource('fake'))
sinkDict = piEval(n, sourceDict)
for x in sinkDict['__0__'].values:
self.assertTrue(x)
print("End of LE Test 2")
def test_SMASeries(self):
with Network() as n:
sourceDict = {'fake': randomWalkSeries()}
df = seriesSource('fake')
sma(df, window=10, name="test")
toComp = sourceDict['fake'].rolling(window=10).mean()
sinkDict = piEval(n, sourceDict)
toComp2 = sinkDict['test']
self.assertIsNone(pd.testing.assert_series_equal(toComp, toComp2))
n.report()
def test_SUMSeries(self):
#Basic sumtest for Series object
with Network() as n:
sourceDict = {'fake': randomWalkSeries()} # here series are loaded
df = seriesSource('fake')
sum(df, name="test")
toComp = sourceDict['fake'].sum()
sinkDict = piEval(n, sourceDict)
toComp2 = sinkDict['test']
self.assertEqual(toComp, toComp2)
n.report()
def test_STDEV_WindowString(self):
with Network() as n:
forex = pd.read_csv("forex.csv")
sourceDict = {'forex': forex} # here series are loaded
df = seriesSource('forex')
stdev(df, window="AskPrice1", name="test")
toComp = sourceDict['forex'].loc[:, "AskPrice1"].std()
sinkDict = piEval(n, sourceDict)
toComp2 = sinkDict['test']
self.assertEqual(toComp, toComp2)
n.report()
def test_STDEV_Window1(self):
with Network() as n:
forex = pd.read_csv("forex.csv")
sourceDict = {'forex': forex} # here series are loaded
df = seriesSource('forex')
stdev(df, window=1, name="test")
toComp = sourceDict['forex'].std(axis=1)
sinkDict = piEval(n, sourceDict)
toComp2 = sinkDict['test']
self.assertIsNone(pd.testing.assert_series_equal(toComp, toComp2))
n.report()
def test_EMADataframe(self):
with Network() as n:
forex = pd.read_csv("forex.csv")
sourceDict = {'forex': forex} # here series are loaded
df = seriesSource('forex')
ema(df, span=10, name="test")
toComp = sourceDict['forex'].ewm(span=10).mean()
sinkDict = piEval(n, sourceDict)
toComp2 = sinkDict['test']
self.assertIsNone(pd.testing.assert_frame_equal(toComp, toComp2))
n.report()
def test_Equal_2(self):
with Network() as n:
sourceDict = {
'fake': randomWalkSeries3()
} # here series are loaded
equal(seriesSource('fake'), seriesSource('fake'))
sinkDict = piEval(n, sourceDict)
for x in sinkDict['__0__'].values:
print(x)
self.assertTrue(x)
print("End of Equal Test")
def test_lessThan(self):
with Network() as n:
sourceDict = {
'fake': randomWalkSeries(),
'fake2': randomWalkSeries2()
} # here series are loaded
lessThan(seriesSource('fake2'), seriesSource('fake'))
sinkDict = piEval(n, sourceDict)
for x in sinkDict['__0__']:
self.assertTrue(x)
print("End of LT Test")
def test_lessorEqual_3(self):
with Network() as n:
sourceDict = {
'fake': randomWalkSeries(),
'fake2': randomWalkSeries2()
} # here series are loaded
lessOrEqual(seriesSource('fake'), seriesSource('fake2'))
sinkDict = piEval(n, sourceDict)
for x in sinkDict['__0__']:
self.assertFalse(x)
print("End of LE Test 3")
def test_SMA_Strategy_MINI(self):
with Network() as n:
forex = pd.read_csv("forex-mini.csv")
sourceDict = {'forex': forex} # here series are loaded
df = seriesSource('forex')
simple_2SMA_Strategy(df,
shortWindow=10,
longWindow=200,
name="test")
sinkDict = piEval(n, sourceDict)
n.report()
def test_LessThanOrEqual_Equal_Series(self):
with Network() as n:
sourceDict = {'fake': randomWalkSeries()} # here series are loaded
a = seriesSource('fake')
b = seriesSource('fake')
lessOrEqual(a, b, name='test')
toComp = pd.Series.le(sourceDict['fake'], sourceDict['fake'])
sinkDict = piEval(n, sourceDict)
toComp2 = sinkDict['test']
self.assertIsNone(pd.testing.assert_series_equal(toComp, toComp2))
n.report()
def testMinRolling(self):
with Network() as n:
sourceDict = {'fake': randomWalkSeries()} # here series are loaded
df = seriesSource('fake')
min(df, name="test", window=5)
toComp = sourceDict['fake']
toCompSeries = pd.Series(toComp).rolling(window=5).min()
sinkDict = piEval(n, sourceDict)
toComp2 = sinkDict['test']
self.assertTrue(
x == y for x, y in zip(toCompSeries.values, toComp2.values))
n.report()
def test_SIMPLE2SMA_TINY_EZ(self):
with Network() as n:
forex = pd.read_csv("forex-tiny-ez.csv")
sourceDict = {'forex': forex}
df = seriesSource('forex')
simple_2SMA_Strategy(df,
shortWindow=1,
longWindow=2,
quantity=1,
name='2SMA')
sinkDict = piEval(n, sourceDict)
n.report()
def test_SUMDataframe(self):
#Basis sumtest for Dataframe object
with Network() as n:
forex = pd.read_csv('forex.csv')
sourceDict = {'forex': forex} # here series are loaded
df = seriesSource('forex')
sum(df, name="test")
toComp = sourceDict['forex'].sum()
sinkDict = piEval(n, sourceDict)
toComp2 = sinkDict['test']
self.assertEqual(toComp.all(), toComp2.all())
n.report()
def test_DELAYSeries(self):
#Basic delay test for Series
with Network() as n:
sourceDict = {'fake': randomWalkSeries()} # here series are loaded
df = seriesSource('fake')
sample = 3
delay(df, sample, name='test')
toComp = sourceDict['fake'].shift(periods=sample)
sinkDict = piEval(n, sourceDict)
toComp2 = sinkDict['test']
self.assertIsNone(pd.testing.assert_series_equal(toComp, toComp2))
n.report()
def test_SMADataframe(self):
with Network() as n:
forex = pd.read_csv("forex.csv")
sourceDict = {'forex': forex} # here series are loaded
df = seriesSource('forex')
sma(df, window=5, name="test")
toComp = sourceDict['forex'].rolling(window=5,
closed="right").mean()
sinkDict = piEval(n, sourceDict)
toComp2 = sinkDict['test']
self.assertIsNone(pd.testing.assert_frame_equal(toComp, toComp2))
n.report()
def test_LessThanOrEqual_Equal_DataFrame(self):
with Network() as n:
forex = pd.read_csv('forex.csv')
sourceDict = {'forex': forex} # here series are loaded
a = seriesSource('forex')
b = seriesSource('forex')
lessOrEqual(a, b, name='test')
toComp = pd.Series.le(sourceDict['forex'], sourceDict['forex'])
sinkDict = piEval(n, sourceDict)
toComp2 = sinkDict['test']
self.assertIsNone(pd.testing.assert_frame_equal(toComp, toComp2))
n.report()
def test_LessThanOrEqual_error(self):
with Network() as n:
sourceDict = {
'fake': randomWalkSeries(),
"fake2": randomWalkDataframe()
} # here series are loaded
a = seriesSource('fake')
b = seriesSource('fake2')
lessOrEqual(a, b, name='test')
sinkDict = piEval(n, sourceDict)
self.assertRaises(Exception, piEval, n, sourceDict)
n.report()
def test_DELAYDataframe(self):
#Basic delay test for Dataframe
with Network() as n:
forex = pd.read_csv("forex.csv")
sourceDict = {'forex': forex} # here series are loaded
df = seriesSource('forex')
sample = 3
delay(df, sample, name='test')
toComp = sourceDict['forex'].shift(periods=sample)
sinkDict = piEval(n, sourceDict)
toComp2 = sinkDict['test']
self.assertIsNone(pd.testing.assert_frame_equal(toComp, toComp2))
n.report()
def test_GetColString(self):
#Checks to make sure inputting a single string returns the correct column as a pd.series object
with Network() as n:
forex = pd.read_csv("forex.csv")
sourceDict = {'forex': forex} # here series are loaded
df = seriesSource('forex')
columnNames = "DateTime"
getColumns(df, columnNames, name='test')
toComp = sourceDict['forex'].get(columnNames)
sinkDict = piEval(n, sourceDict)
toComp2 = sinkDict['test']
self.assertIsNone(pd.testing.assert_series_equal(toComp, toComp2))
n.report()