def on_preparing_data(self, symbol, data, context: {}):
"""循环对 `linear_angle` 指定的日期区间，调用 ``talib.LINEARREG_ANGLE`` ，生成角度值。
所有计算得到的值取绝对值，对没有值的内容填充 `99` （以示角度超大）。
取到的所有值会放在 ``context`` 中，key值为 `angle` ；
根据 `max_linear_angle` 设置的值对每隔日期区间的角度制取平均值或中位数。
取得的值会放在 ``context`` 中，key值为 `max_angles` ；
"""
angles = []
max_angles = {}
for angle in self.linear_angle:
col = 'angle_{}'.format(angle)
angles.append(col)
data[col] = talib.LINEARREG_ANGLE(data['close'],
angle).abs().fillna(99) # 日均线角度
if self.max_linear_angle == "MEAN":
v = talib.LINEARREG_ANGLE(data['close'],
angle).abs().mean() # 日均线角度平均值
else:
v = talib.LINEARREG_ANGLE(data['close'],
angle).abs().median() # 日均线角度中位数
if np.isnan(v):
raise ValueError('{} LINEARREG_ANGLE is NaN'.format(angle))
max_angles[col] = v
context['angles'] = angles
context['max_angles'] = max_angles
def test_LINEARREG_ANGLE(self):
self.env.add_operator('linearreg', {
'operator': OperatorLINEARREG_ANGLE,
})
string = 'linearreg(14, open)'
gene = self.env.parse_string(string)
self.assertRaises(IndexError, gene.eval, self.env, self.dates[12], self.dates[-1])
df = gene.eval(self.env, self.dates[13], self.dates[14])
ser0, ser1 = df.iloc[0], df.iloc[1]
o = self.env.get_data_value('open').values
res0, res1, res = [], [], []
for i in df.columns:
res0.append(talib.LINEARREG_ANGLE(o[:14, i], timeperiod=14)[-1] == ser0[i])
res1.append(talib.LINEARREG_ANGLE(o[1:14+1, i], timeperiod=14)[-1] == ser1[i])
res.append(talib.LINEARREG_ANGLE(o[:14+1, i], timeperiod=14)[-1] == ser1[i])
self.assertTrue(all(res0) and all(res1) and all(res))
def trend(df, target_list):
for target in target_list:
trend = talib.LINEARREG_ANGLE(df[target], timeperiod=10)
df['TREND_' + target] = trend
return df
def eval(self, environment, gene, date1, date2):
timeperiod = (gene.next_value(environment, date1, date2))
date1 = environment.shift_date(date1, -(timeperiod - 1), -1)
df = gene.next_value(environment, date1, date2)
res = df.apply(lambda x: pd.Series(talib.LINEARREG_ANGLE(
x.values, timeperiod=timeperiod),
index=df.index))
return res.iloc[timeperiod - 1:]
def LINEARREG_ANGLE(close, timeperiod=14):
''' Linear Regression Angle 线性回归的角度
分组: Statistic Functions 统计函数
简介:
real = LINEARREG_ANGLE(close, timeperiod=14)
'''
return talib.LINEARREG_ANGLE(close, timeperiod)
def get_angle(ss, p=4, ang_type='degree'):
fss = np.nan_to_num(ss, 0)
scale = (max(fss) - min(fss)) / 2 * 1.618
nfss = fss / scale
# pdb.set_trace()
angle = talib.LINEARREG_ANGLE(nfss, timeperiod=p)
if ang_type == 'degree':
return np.arctan(angle) / np.pi * 180
elif ang_type == 'pi':
return np.arctan(angle)
else:
return angle
def _talib_LINEARREG_ANGLE(data, n):
if n <= 1:
value = np.zeros_like(data)
else:
try:
value = talib.LINEARREG_ANGLE(data, timeperiod=n)
ss = pd.Series(value).ffill().fillna(0)
value = ss.values
except Exception as e:
raise Exception("[_talib_LINEARREG_ANGLE]", e)
# print("[WARNING] _talib_LINEARREG_ANGLE: {}".format(e.args[0]))
value = np.zeros_like(data)
return value.astype(float)
def test_linreg_angle(self):
result = self.overlap.linreg(self.close, angle=True)
self.assertIsInstance(result, Series)
self.assertEqual(result.name, 'LRa_14')
try:
expected = tal.LINEARREG_ANGLE(self.close)
pdt.assert_series_equal(result, expected, check_names=False)
except AssertionError as ae:
try:
corr = pandas_ta.utils.df_error_analysis(result, expected, col=CORRELATION)
self.assertGreater(corr, CORRELATION_THRESHOLD)
except Exception as ex:
error_analysis(result, CORRELATION, ex)
def getStatFunctions(df):
high = df['High']
low = df['Low']
close = df['Close']
open = df['Open']
volume = df['Volume']
df['BETA'] = ta.BETA(high, low, timeperiod=5)
df['CORREL'] = ta.CORREL(high, low, timeperiod=30)
df['LINREG'] = ta.LINEARREG(close, timeperiod=14)
df['LINREGANGLE'] = ta.LINEARREG_ANGLE(close, timeperiod=14)
df['LINREGINTERCEPT'] = ta.LINEARREG_INTERCEPT(close, timeperiod=14)
df['LINREGSLOPE'] = ta.LINEARREG_SLOPE(close, timeperiod=14)
df['STDDEV'] = ta.STDDEV(close, timeperiod=5, nbdev=1)
df['TSF'] = ta.TSF(close, timeperiod=14)
df['VAR'] = ta.VAR(close, timeperiod=5, nbdev=1)
def linearreg_angle(client, symbol, timeframe="6m", closecol="close", period=14):
"""This will return a dataframe of linear regression angle for the given symbol across
the given timeframe
Args:
client (pyEX.Client): Client
symbol (string): Ticker
timeframe (string): timeframe to use, for pyEX.chart
closecol (string): column to use to calculate
period (int): period to calculate adx across
Returns:
DataFrame: result
"""
df = client.chartDF(symbol, timeframe)
linearreg = t.LINEARREG_ANGLE(df[closecol].values, period)
return pd.DataFrame({closecol: df[closecol].values, "lineearreg_angle": linearreg})
def linearreg_angle(candles: np.ndarray, period: int = 14, source_type: str = "close", sequential: bool = False) -> \
Union[float, np.ndarray]:
"""
LINEARREG_ANGLE - Linear Regression Angle
:param candles: np.ndarray
:param period: int - default: 14
:param source_type: str - default: "close"
:param sequential: bool - default=False
:return: float | np.ndarray
"""
candles = slice_candles(candles, sequential)
source = get_candle_source(candles, source_type=source_type)
res = talib.LINEARREG_ANGLE(source, timeperiod=period)
return res if sequential else res[-1]
def linearreg_angle(candles: np.ndarray, period: int = 14, source_type: str = "close", sequential: bool = False) -> \
Union[float, np.ndarray]:
"""
LINEARREG_ANGLE - Linear Regression Angle
:param candles: np.ndarray
:param period: int - default: 14
:param source_type: str - default: "close"
:param sequential: bool - default=False
:return: float | np.ndarray
"""
warmup_candles_num = get_config('env.data.warmup_candles_num', 240)
if not sequential and len(candles) > warmup_candles_num:
candles = candles[-warmup_candles_num:]
source = get_candle_source(candles, source_type=source_type)
res = talib.LINEARREG_ANGLE(source, timeperiod=period)
return res if sequential else res[-1]
def statistic_process(event):
print(event.widget.get())
statistic = event.widget.get()
upperband, middleband, lowerband = ta.BBANDS(close,
timeperiod=5,
nbdevup=2,
nbdevdn=2,
matype=0)
fig, axes = plt.subplots(2, 1, sharex=True)
ax1, ax2 = axes[0], axes[1]
axes[0].plot(close, 'rd-', markersize=3)
axes[0].plot(upperband, 'y-')
axes[0].plot(middleband, 'b-')
axes[0].plot(lowerband, 'y-')
axes[0].set_title(statistic, fontproperties='SimHei')
if statistic == '线性回归':
real = ta.LINEARREG(close, timeperiod=14)
axes[1].plot(real, 'r-')
elif statistic == '线性回归角度':
real = ta.LINEARREG_ANGLE(close, timeperiod=14)
axes[1].plot(real, 'r-')
elif statistic == '线性回归截距':
real = ta.LINEARREG_INTERCEPT(close, timeperiod=14)
axes[1].plot(real, 'r-')
elif statistic == '线性回归斜率':
real = ta.LINEARREG_SLOPE(close, timeperiod=14)
axes[1].plot(real, 'r-')
elif statistic == '标准差':
real = ta.STDDEV(close, timeperiod=5, nbdev=1)
axes[1].plot(real, 'r-')
elif statistic == '时间序列预测':
real = ta.TSF(close, timeperiod=14)
axes[1].plot(real, 'r-')
elif statistic == '方差':
real = ta.VAR(close, timeperiod=5, nbdev=1)
axes[1].plot(real, 'r-')
plt.show()
def Stat_Function(dataframe):
#Statistic Functions
#BETA - Beta
df[f'{ratio}_BETA'] = talib.BETA(High, Low, timeperiod=5)
#CORREL - Pearson's Correlation Coefficient (r)
df[f'{ratio}_CORREL'] = talib.CORREL(High, Low, timeperiod=30)
#LINEARREG - Linear Regression
df[f'{ratio}_LINEARREG'] = talib.LINEARREG(Close, timeperiod=14)
#LINEARREG_ANGLE - Linear Regression Angle
df[f'{ratio}_LINEARREG_ANGLE'] = talib.LINEARREG_ANGLE(Close, timeperiod=14)
#LINEARREG_INTERCEPT - Linear Regression Intercept
df[f'{ratio}_LINEARREG_INTERCEPT'] = talib.LINEARREG_INTERCEPT(Close, timeperiod=14)
#LINEARREG_SLOPE - Linear Regression Slope
df[f'{ratio}_LINEARREG_SLOPE'] = talib.LINEARREG_SLOPE(Close, timeperiod=14)
#STDDEV - Standard Deviation
df[f'{ratio}_STDDEV'] = talib.STDDEV(Close, timeperiod=5, nbdev=1)
#TSF - Time Series Forecast
df[f'{ratio}_TSF'] = talib.TSF(Close, timeperiod=14)
#VAR - Variance
df[f'{ratio}_VAR'] = talib.VAR(Close, timeperiod=5, nbdev=1)
return
def calculate(self, para):
self.t = self.inputdata[:, 0]
self.op = self.inputdata[:, 1]
self.high = self.inputdata[:, 2]
self.low = self.inputdata[:, 3]
self.close = self.inputdata[:, 4]
#adjusted close
self.close1 = self.inputdata[:, 5]
self.volume = self.inputdata[:, 6]
#Overlap study
#Overlap Studies
#Overlap Studies
if para is 'BBANDS': #Bollinger Bands
upperband, middleband, lowerband = ta.BBANDS(self.close,
timeperiod=self.tp,
nbdevup=2,
nbdevdn=2,
matype=0)
self.output = [upperband, middleband, lowerband]
elif para is 'DEMA': #Double Exponential Moving Average
self.output = ta.DEMA(self.close, timeperiod=self.tp)
elif para is 'EMA': #Exponential Moving Average
self.output = ta.EMA(self.close, timeperiod=self.tp)
elif para is 'HT_TRENDLINE': #Hilbert Transform - Instantaneous Trendline
self.output = ta.HT_TRENDLINE(self.close)
elif para is 'KAMA': #Kaufman Adaptive Moving Average
self.output = ta.KAMA(self.close, timeperiod=self.tp)
elif para is 'MA': #Moving average
self.output = ta.MA(self.close, timeperiod=self.tp, matype=0)
elif para is 'MAMA': #MESA Adaptive Moving Average
mama, fama = ta.MAMA(self.close, fastlimit=0, slowlimit=0)
elif para is 'MAVP': #Moving average with variable period
self.output = ta.MAVP(self.close,
periods=10,
minperiod=self.tp,
maxperiod=self.tp1,
matype=0)
elif para is 'MIDPOINT': #MidPoint over period
self.output = ta.MIDPOINT(self.close, timeperiod=self.tp)
elif para is 'MIDPRICE': #Midpoint Price over period
self.output = ta.MIDPRICE(self.high, self.low, timeperiod=self.tp)
elif para is 'SAR': #Parabolic SAR
self.output = ta.SAR(self.high,
self.low,
acceleration=0,
maximum=0)
elif para is 'SAREXT': #Parabolic SAR - Extended
self.output = ta.SAREXT(self.high,
self.low,
startvalue=0,
offsetonreverse=0,
accelerationinitlong=0,
accelerationlong=0,
accelerationmaxlong=0,
accelerationinitshort=0,
accelerationshort=0,
accelerationmaxshort=0)
elif para is 'SMA': #Simple Moving Average
self.output = ta.SMA(self.close, timeperiod=self.tp)
elif para is 'T3': #Triple Exponential Moving Average (T3)
self.output = ta.T3(self.close, timeperiod=self.tp, vfactor=0)
elif para is 'TEMA': #Triple Exponential Moving Average
self.output = ta.TEMA(self.close, timeperiod=self.tp)
elif para is 'TRIMA': #Triangular Moving Average
self.output = ta.TRIMA(self.close, timeperiod=self.tp)
elif para is 'WMA': #Weighted Moving Average
self.output = ta.WMA(self.close, timeperiod=self.tp)
#Momentum Indicators
elif para is 'ADX': #Average Directional Movement Index
self.output = ta.ADX(self.high,
self.low,
self.close,
timeperiod=self.tp)
elif para is 'ADXR': #Average Directional Movement Index Rating
self.output = ta.ADXR(self.high,
self.low,
self.close,
timeperiod=self.tp)
elif para is 'APO': #Absolute Price Oscillator
self.output = ta.APO(self.close,
fastperiod=12,
slowperiod=26,
matype=0)
elif para is 'AROON': #Aroon
aroondown, aroonup = ta.AROON(self.high,
self.low,
timeperiod=self.tp)
self.output = [aroondown, aroonup]
elif para is 'AROONOSC': #Aroon Oscillator
self.output = ta.AROONOSC(self.high, self.low, timeperiod=self.tp)
elif para is 'BOP': #Balance Of Power
self.output = ta.BOP(self.op, self.high, self.low, self.close)
elif para is 'CCI': #Commodity Channel Index
self.output = ta.CCI(self.high,
self.low,
self.close,
timeperiod=self.tp)
elif para is 'CMO': #Chande Momentum Oscillator
self.output = ta.CMO(self.close, timeperiod=self.tp)
elif para is 'DX': #Directional Movement Index
self.output = ta.DX(self.high,
self.low,
self.close,
timeperiod=self.tp)
elif para is 'MACD': #Moving Average Convergence/Divergence
macd, macdsignal, macdhist = ta.MACD(self.close,
fastperiod=12,
slowperiod=26,
signalperiod=9)
self.output = [macd, macdsignal, macdhist]
elif para is 'MACDEXT': #MACD with controllable MA type
macd, macdsignal, macdhist = ta.MACDEXT(self.close,
fastperiod=12,
fastmatype=0,
slowperiod=26,
slowmatype=0,
signalperiod=9,
signalmatype=0)
self.output = [macd, macdsignal, macdhist]
elif para is 'MACDFIX': #Moving Average Convergence/Divergence Fix 12/26
macd, macdsignal, macdhist = ta.MACDFIX(self.close, signalperiod=9)
self.output = [macd, macdsignal, macdhist]
elif para is 'MFI': #Money Flow Index
self.output = ta.MFI(self.high,
self.low,
self.close,
self.volume,
timeperiod=self.tp)
elif para is 'MINUS_DI': #Minus Directional Indicator
self.output = ta.MINUS_DI(self.high,
self.low,
self.close,
timeperiod=self.tp)
elif para is 'MINUS_DM': #Minus Directional Movement
self.output = ta.MINUS_DM(self.high, self.low, timeperiod=self.tp)
elif para is 'MOM': #Momentum
self.output = ta.MOM(self.close, timeperiod=10)
elif para is 'PLUS_DI': #Plus Directional Indicator
self.output = ta.PLUS_DI(self.high,
self.low,
self.close,
timeperiod=self.tp)
elif para is 'PLUS_DM': #Plus Directional Movement
self.output = ta.PLUS_DM(self.high, self.low, timeperiod=self.tp)
elif para is 'PPO': #Percentage Price Oscillator
self.output = ta.PPO(self.close,
fastperiod=12,
slowperiod=26,
matype=0)
elif para is 'ROC': #Rate of change : ((price/prevPrice)-1)*100
self.output = ta.ROC(self.close, timeperiod=10)
elif para is 'ROCP': #Rate of change Percentage: (price-prevPrice)/prevPrice
self.output = ta.ROCP(self.close, timeperiod=10)
elif para is 'ROCR': #Rate of change ratio: (price/prevPrice)
self.output = ta.ROCR(self.close, timeperiod=10)
elif para is 'ROCR100': #Rate of change ratio 100 scale: (price/prevPrice)*100
self.output = ta.ROCR100(self.close, timeperiod=10)
elif para is 'RSI': #Relative Strength Index
self.output = ta.RSI(self.close, timeperiod=self.tp)
elif para is 'STOCH': #Stochastic
slowk, slowd = ta.STOCH(self.high,
self.low,
self.close,
fastk_period=5,
slowk_period=3,
slowk_matype=0,
slowd_period=3,
slowd_matype=0)
self.output = [slowk, slowd]
elif para is 'STOCHF': #Stochastic Fast
fastk, fastd = ta.STOCHF(self.high,
self.low,
self.close,
fastk_period=5,
fastd_period=3,
fastd_matype=0)
self.output = [fastk, fastd]
elif para is 'STOCHRSI': #Stochastic Relative Strength Index
fastk, fastd = ta.STOCHRSI(self.close,
timeperiod=self.tp,
fastk_period=5,
fastd_period=3,
fastd_matype=0)
self.output = [fastk, fastd]
elif para is 'TRIX': #1-day Rate-Of-Change (ROC) of a Triple Smooth EMA
self.output = ta.TRIX(self.close, timeperiod=self.tp)
elif para is 'ULTOSC': #Ultimate Oscillator
self.output = ta.ULTOSC(self.high,
self.low,
self.close,
timeperiod1=self.tp,
timeperiod2=self.tp1,
timeperiod3=self.tp2)
elif para is 'WILLR': #Williams' %R
self.output = ta.WILLR(self.high,
self.low,
self.close,
timeperiod=self.tp)
# Volume Indicators : #
elif para is 'AD': #Chaikin A/D Line
self.output = ta.AD(self.high, self.low, self.close, self.volume)
elif para is 'ADOSC': #Chaikin A/D Oscillator
self.output = ta.ADOSC(self.high,
self.low,
self.close,
self.volume,
fastperiod=3,
slowperiod=10)
elif para is 'OBV': #On Balance Volume
self.output = ta.OBV(self.close, self.volume)
# Volatility Indicators: #
elif para is 'ATR': #Average True Range
self.output = ta.ATR(self.high,
self.low,
self.close,
timeperiod=self.tp)
elif para is 'NATR': #Normalized Average True Range
self.output = ta.NATR(self.high,
self.low,
self.close,
timeperiod=self.tp)
elif para is 'TRANGE': #True Range
self.output = ta.TRANGE(self.high, self.low, self.close)
#Price Transform : #
elif para is 'AVGPRICE': #Average Price
self.output = ta.AVGPRICE(self.op, self.high, self.low, self.close)
elif para is 'MEDPRICE': #Median Price
self.output = ta.MEDPRICE(self.high, self.low)
elif para is 'TYPPRICE': #Typical Price
self.output = ta.TYPPRICE(self.high, self.low, self.close)
elif para is 'WCLPRICE': #Weighted Close Price
self.output = ta.WCLPRICE(self.high, self.low, self.close)
#Cycle Indicators : #
elif para is 'HT_DCPERIOD': #Hilbert Transform - Dominant Cycle Period
self.output = ta.HT_DCPERIOD(self.close)
elif para is 'HT_DCPHASE': #Hilbert Transform - Dominant Cycle Phase
self.output = ta.HT_DCPHASE(self.close)
elif para is 'HT_PHASOR': #Hilbert Transform - Phasor Components
inphase, quadrature = ta.HT_PHASOR(self.close)
self.output = [inphase, quadrature]
elif para is 'HT_SINE': #Hilbert Transform - SineWave #2
sine, leadsine = ta.HT_SINE(self.close)
self.output = [sine, leadsine]
elif para is 'HT_TRENDMODE': #Hilbert Transform - Trend vs Cycle Mode
self.integer = ta.HT_TRENDMODE(self.close)
#Pattern Recognition : #
elif para is 'CDL2CROWS': #Two Crows
self.integer = ta.CDL2CROWS(self.op, self.high, self.low,
self.close)
elif para is 'CDL3BLACKCROWS': #Three Black Crows
self.integer = ta.CDL3BLACKCROWS(self.op, self.high, self.low,
self.close)
elif para is 'CDL3INSIDE': #Three Inside Up/Down
self.integer = ta.CDL3INSIDE(self.op, self.high, self.low,
self.close)
elif para is 'CDL3LINESTRIKE': #Three-Line Strike
self.integer = ta.CDL3LINESTRIKE(self.op, self.high, self.low,
self.close)
elif para is 'CDL3OUTSIDE': #Three Outside Up/Down
self.integer = ta.CDL3OUTSIDE(self.op, self.high, self.low,
self.close)
elif para is 'CDL3STARSINSOUTH': #Three Stars In The South
self.integer = ta.CDL3STARSINSOUTH(self.op, self.high, self.low,
self.close)
elif para is 'CDL3WHITESOLDIERS': #Three Advancing White Soldiers
self.integer = ta.CDL3WHITESOLDIERS(self.op, self.high, self.low,
self.close)
elif para is 'CDLABANDONEDBABY': #Abandoned Baby
self.integer = ta.CDLABANDONEDBABY(self.op,
self.high,
self.low,
self.close,
penetration=0)
elif para is 'CDLBELTHOLD': #Belt-hold
self.integer = ta.CDLBELTHOLD(self.op, self.high, self.low,
self.close)
elif para is 'CDLBREAKAWAY': #Breakaway
self.integer = ta.CDLBREAKAWAY(self.op, self.high, self.low,
self.close)
elif para is 'CDLCLOSINGMARUBOZU': #Closing Marubozu
self.integer = ta.CDLCLOSINGMARUBOZU(self.op, self.high, self.low,
self.close)
elif para is 'CDLCONCEALBABYSWALL': #Concealing Baby Swallow
self.integer = ta.CDLCONCEALBABYSWALL(self.op, self.high, self.low,
self.close)
elif para is 'CDLCOUNTERATTACK': #Counterattack
self.integer = ta.CDLCOUNTERATTACK(self.op, self.high, self.low,
self.close)
elif para is 'CDLDARKCLOUDCOVER': #Dark Cloud Cover
self.integer = ta.CDLDARKCLOUDCOVER(self.op,
self.high,
self.low,
self.close,
penetration=0)
elif para is 'CDLDOJI': #Doji
self.integer = ta.CDLDOJI(self.op, self.high, self.low, self.close)
elif para is 'CDLDOJISTAR': #Doji Star
self.integer = ta.CDLDOJISTAR(self.op, self.high, self.low,
self.close)
elif para is 'CDLDRAGONFLYDOJI': #Dragonfly Doji
self.integer = ta.CDLDRAGONFLYDOJI(self.op, self.high, self.low,
self.close)
elif para is 'CDLENGULFING': #Engulfing Pattern
self.integer = ta.CDLENGULFING(self.op, self.high, self.low,
self.close)
elif para is 'CDLEVENINGDOJISTAR': #Evening Doji Star
self.integer = ta.CDLEVENINGDOJISTAR(self.op,
self.high,
self.low,
self.close,
penetration=0)
elif para is 'CDLEVENINGSTAR': #Evening Star
self.integer = ta.CDLEVENINGSTAR(self.op,
self.high,
self.low,
self.close,
penetration=0)
elif para is 'CDLGAPSIDESIDEWHITE': #Up/Down-gap side-by-side white lines
self.integer = ta.CDLGAPSIDESIDEWHITE(self.op, self.high, self.low,
self.close)
elif para is 'CDLGRAVESTONEDOJI': #Gravestone Doji
self.integer = ta.CDLGRAVESTONEDOJI(self.op, self.high, self.low,
self.close)
elif para is 'CDLHAMMER': #Hammer
self.integer = ta.CDLHAMMER(self.op, self.high, self.low,
self.close)
elif para is 'CDLHANGINGMAN': #Hanging Man
self.integer = ta.CDLHANGINGMAN(self.op, self.high, self.low,
self.close)
elif para is 'CDLHARAMI': #Harami Pattern
self.integer = ta.CDLHARAMI(self.op, self.high, self.low,
self.close)
elif para is 'CDLHARAMICROSS': #Harami Cross Pattern
self.integer = ta.CDLHARAMICROSS(self.op, self.high, self.low,
self.close)
elif para is 'CDLHIGHWAVE': #High-Wave Candle
self.integer = ta.CDLHIGHWAVE(self.op, self.high, self.low,
self.close)
elif para is 'CDLHIKKAKE': #Hikkake Pattern
self.integer = ta.CDLHIKKAKE(self.op, self.high, self.low,
self.close)
elif para is 'CDLHIKKAKEMOD': #Modified Hikkake Pattern
self.integer = ta.CDLHIKKAKEMOD(self.op, self.high, self.low,
self.close)
elif para is 'CDLHOMINGPIGEON': #Homing Pigeon
self.integer = ta.CDLHOMINGPIGEON(self.op, self.high, self.low,
self.close)
elif para is 'CDLIDENTICAL3CROWS': #Identical Three Crows
self.integer = ta.CDLIDENTICAL3CROWS(self.op, self.high, self.low,
self.close)
elif para is 'CDLINNECK': #In-Neck Pattern
self.integer = ta.CDLINNECK(self.op, self.high, self.low,
self.close)
elif para is 'CDLINVERTEDHAMMER': #Inverted Hammer
self.integer = ta.CDLINVERTEDHAMMER(self.op, self.high, self.low,
self.close)
elif para is 'CDLKICKING': #Kicking
self.integer = ta.CDLKICKING(self.op, self.high, self.low,
self.close)
elif para is 'CDLKICKINGBYLENGTH': #Kicking - bull/bear determined by the longer marubozu
self.integer = ta.CDLKICKINGBYLENGTH(self.op, self.high, self.low,
self.close)
elif para is 'CDLLADDERBOTTOM': #Ladder Bottom
self.integer = ta.CDLLADDERBOTTOM(self.op, self.high, self.low,
self.close)
elif para is 'CDLLONGLEGGEDDOJI': #Long Legged Doji
self.integer = ta.CDLLONGLEGGEDDOJI(self.op, self.high, self.low,
self.close)
elif para is 'CDLLONGLINE': #Long Line Candle
self.integer = ta.CDLLONGLINE(self.op, self.high, self.low,
self.close)
elif para is 'CDLMARUBOZU': #Marubozu
self.integer = ta.CDLMARUBOZU(self.op, self.high, self.low,
self.close)
elif para is 'CDLMATCHINGLOW': #Matching Low
self.integer = ta.CDLMATCHINGLOW(self.op, self.high, self.low,
self.close)
elif para is 'CDLMATHOLD': #Mat Hold
self.integer = ta.CDLMATHOLD(self.op,
self.high,
self.low,
self.close,
penetration=0)
elif para is 'CDLMORNINGDOJISTAR': #Morning Doji Star
self.integer = ta.CDLMORNINGDOJISTAR(self.op,
self.high,
self.low,
self.close,
penetration=0)
elif para is 'CDLMORNINGSTAR': #Morning Star
self.integer = ta.CDLMORNINGSTAR(self.op,
self.high,
self.low,
self.close,
penetration=0)
elif para is 'CDLONNECK': #On-Neck Pattern
self.integer = ta.CDLONNECK(self.op, self.high, self.low,
self.close)
elif para is 'CDLPIERCING': #Piercing Pattern
self.integer = ta.CDLPIERCING(self.op, self.high, self.low,
self.close)
elif para is 'CDLRICKSHAWMAN': #Rickshaw Man
self.integer = ta.CDLRICKSHAWMAN(self.op, self.high, self.low,
self.close)
elif para is 'CDLRISEFALL3METHODS': #Rising/Falling Three Methods
self.integer = ta.CDLRISEFALL3METHODS(self.op, self.high, self.low,
self.close)
elif para is 'CDLSEPARATINGLINES': #Separating Lines
self.integer = ta.CDLSEPARATINGLINES(self.op, self.high, self.low,
self.close)
elif para is 'CDLSHOOTINGSTAR': #Shooting Star
self.integer = ta.CDLSHOOTINGSTAR(self.op, self.high, self.low,
self.close)
elif para is 'CDLSHORTLINE': #Short Line Candle
self.integer = ta.CDLSHORTLINE(self.op, self.high, self.low,
self.close)
elif para is 'CDLSPINNINGTOP': #Spinning Top
self.integer = ta.CDLSPINNINGTOP(self.op, self.high, self.low,
self.close)
elif para is 'CDLSTALLEDPATTERN': #Stalled Pattern
self.integer = ta.CDLSTALLEDPATTERN(self.op, self.high, self.low,
self.close)
elif para is 'CDLSTICKSANDWICH': #Stick Sandwich
self.integer = ta.CDLSTICKSANDWICH(self.op, self.high, self.low,
self.close)
elif para is 'CDLTAKURI': #Takuri (Dragonfly Doji with very long lower shadow)
self.integer = ta.CDLTAKURI(self.op, self.high, self.low,
self.close)
elif para is 'CDLTASUKIGAP': #Tasuki Gap
self.integer = ta.CDLTASUKIGAP(self.op, self.high, self.low,
self.close)
elif para is 'CDLTHRUSTING': #Thrusting Pattern
self.integer = ta.CDLTHRUSTING(self.op, self.high, self.low,
self.close)
elif para is 'CDLTRISTAR': #Tristar Pattern
self.integer = ta.CDLTRISTAR(self.op, self.high, self.low,
self.close)
elif para is 'CDLUNIQUE3RIVER': #Unique 3 River
self.integer = ta.CDLUNIQUE3RIVER(self.op, self.high, self.low,
self.close)
elif para is 'CDLUPSIDEGAP2CROWS': #Upside Gap Two Crows
self.integer = ta.CDLUPSIDEGAP2CROWS(self.op, self.high, self.low,
self.close)
elif para is 'CDLXSIDEGAP3METHODS': #Upside/Downside Gap Three Methods
self.integer = ta.CDLXSIDEGAP3METHODS(self.op, self.high, self.low,
self.close)
#Statistic Functions : #
elif para is 'BETA': #Beta
self.output = ta.BETA(self.high, self.low, timeperiod=5)
elif para is 'CORREL': #Pearson's Correlation Coefficient (r)
self.output = ta.CORREL(self.high, self.low, timeperiod=self.tp)
elif para is 'LINEARREG': #Linear Regression
self.output = ta.LINEARREG(self.close, timeperiod=self.tp)
elif para is 'LINEARREG_ANGLE': #Linear Regression Angle
self.output = ta.LINEARREG_ANGLE(self.close, timeperiod=self.tp)
elif para is 'LINEARREG_INTERCEPT': #Linear Regression Intercept
self.output = ta.LINEARREG_INTERCEPT(self.close,
timeperiod=self.tp)
elif para is 'LINEARREG_SLOPE': #Linear Regression Slope
self.output = ta.LINEARREG_SLOPE(self.close, timeperiod=self.tp)
elif para is 'STDDEV': #Standard Deviation
self.output = ta.STDDEV(self.close, timeperiod=5, nbdev=1)
elif para is 'TSF': #Time Series Forecast
self.output = ta.TSF(self.close, timeperiod=self.tp)
elif para is 'VAR': #Variance
self.output = ta.VAR(self.close, timeperiod=5, nbdev=1)
else:
print('You issued command:' + para)
def TALIB_LINEARREG_ANGLE(close, timeperiod=14):
'''00349,2,1'''
return talib.LINEARREG_ANGLE(close, timeperiod)
df['ATR3'] = abs(
np.array(df['Low'].shift(1)) - np.array(df['Adj Close'].shift(1)))
df['AverageTrueRange'] = df[['ATR1', 'ATR2', 'ATR3']].max(axis=1)
# df['EMA']=pd.Series(pd.ewma(df['Adj Close'], span = n, min_periods = n - 1))
# Statistic Functions
df['Beta'] = ta.BETA(np.array(df['High'].shift(1)),
np.array(df['Low'].shift(1)),
timeperiod=n)
df['CORREL'] = ta.CORREL(np.array(df['High'].shift(1)),
np.array(df['Low'].shift(1)),
timeperiod=n)
df['LINEARREG'] = ta.LINEARREG(np.array(df['Adj Close'].shift(1)),
timeperiod=n)
df['LINEARREG_ANGLE'] = ta.LINEARREG_ANGLE(np.array(df['Adj Close'].shift(1)),
timeperiod=n)
df['LINEARREG_INTERCEPT'] = ta.LINEARREG_INTERCEPT(np.array(
df['Adj Close'].shift(1)),
timeperiod=n)
df['LINEARREG_SLOPE'] = ta.LINEARREG_SLOPE(np.array(df['Adj Close'].shift(1)),
timeperiod=n)
df['STDDEV'] = ta.STDDEV(np.array(df['Adj Close'].shift(1)),
timeperiod=n,
nbdev=1)
df['Time Series Forecast'] = ta.TSF(np.array(df['Adj Close'].shift(1)),
timeperiod=n)
df['VAR'] = ta.VAR(np.array(df['Adj Close'].shift(1)), timeperiod=n, nbdev=1)
# Overlap Studies Functions
df['upperband'], df['middleband'], df['lowerband'] = ta.BBANDS(np.array(
df['Adj Close'].shift(1)),
def LINEARREG_ANGLE(data, **kwargs):
_check_talib_presence()
prices = _extract_series(data)
return talib.LINEARREG_ANGLE(prices, **kwargs)
def extract_linear_reg(data):
data['linear_reg_angle'] = ta.LINEARREG_ANGLE(data['close'].values,
timeperiod=short_period)
data['linear_reg_slope'] = ta.LINEARREG_SLOPE(data['close'].values,
timeperiod=short_period)
return data
def add_ta_features(df, ta_settings):
"""Add technial analysis features from typical financial dataset that
typically include columns such as "open", "high", "low", "price" and
"volume".
http://mrjbq7.github.io/ta-lib/
Args:
df(pandas.DataFrame): original DataFrame.
ta_settings(dict): configuration.
Returns:
pandas.DataFrame: DataFrame with new features included.
"""
open = df['open']
high = df['high']
low = df['low']
close = df['price']
volume = df['volume']
if ta_settings['overlap']:
df['ta_overlap_bbands_upper'], df['ta_overlap_bbands_middle'], df[
'ta_overlap_bbands_lower'] = ta.BBANDS(close,
timeperiod=5,
nbdevup=2,
nbdevdn=2,
matype=0)
df['ta_overlap_dema'] = ta.DEMA(
close, timeperiod=15) # NOTE: Changed to avoid a lot of Nan values
df['ta_overlap_ema'] = ta.EMA(close, timeperiod=30)
df['ta_overlap_kama'] = ta.KAMA(close, timeperiod=30)
df['ta_overlap_ma'] = ta.MA(close, timeperiod=30, matype=0)
df['ta_overlap_mama_mama'], df['ta_overlap_mama_fama'] = ta.MAMA(close)
period = np.random.randint(10, 20, size=len(close)).astype(float)
df['ta_overlap_mavp'] = ta.MAVP(close,
period,
minperiod=2,
maxperiod=30,
matype=0)
df['ta_overlap_midpoint'] = ta.MIDPOINT(close, timeperiod=14)
df['ta_overlap_midprice'] = ta.MIDPRICE(high, low, timeperiod=14)
df['ta_overlap_sar'] = ta.SAR(high, low, acceleration=0, maximum=0)
df['ta_overlap_sarext'] = ta.SAREXT(high,
low,
startvalue=0,
offsetonreverse=0,
accelerationinitlong=0,
accelerationlong=0,
accelerationmaxlong=0,
accelerationinitshort=0,
accelerationshort=0,
accelerationmaxshort=0)
df['ta_overlap_sma'] = ta.SMA(close, timeperiod=30)
df['ta_overlap_t3'] = ta.T3(close, timeperiod=5, vfactor=0)
df['ta_overlap_tema'] = ta.TEMA(
close, timeperiod=12) # NOTE: Changed to avoid a lot of Nan values
df['ta_overlap_trima'] = ta.TRIMA(close, timeperiod=30)
df['ta_overlap_wma'] = ta.WMA(close, timeperiod=30)
# NOTE: Commented to avoid a lot of Nan values
# df['ta_overlap_ht_trendline'] = ta.HT_TRENDLINE(close)
if ta_settings['momentum']:
df['ta_momentum_adx'] = ta.ADX(high, low, close, timeperiod=14)
df['ta_momentum_adxr'] = ta.ADXR(high, low, close, timeperiod=14)
df['ta_momentum_apo'] = ta.APO(close,
fastperiod=12,
slowperiod=26,
matype=0)
df['ta_momentum_aroondown'], df['ta_momentum_aroonup'] = ta.AROON(
high, low, timeperiod=14)
df['ta_momentum_aroonosc'] = ta.AROONOSC(high, low, timeperiod=14)
df['ta_momentum_bop'] = ta.BOP(open, high, low, close)
df['ta_momentum_cci'] = ta.CCI(high, low, close, timeperiod=14)
df['ta_momentum_cmo'] = ta.CMO(close, timeperiod=14)
df['ta_momentum_dx'] = ta.DX(high, low, close, timeperiod=14)
df['ta_momentum_macd_macd'], df['ta_momentum_macd_signal'], df[
'ta_momentum_macd_hist'] = ta.MACD(close,
fastperiod=12,
slowperiod=26,
signalperiod=9)
df['ta_momentum_macdext_macd'], df['ta_momentum_macdext_signal'], df[
'ta_momentum_macdext_hist'] = ta.MACDEXT(close,
fastperiod=12,
fastmatype=0,
slowperiod=26,
slowmatype=0,
signalperiod=9,
signalmatype=0)
df['ta_momentum_macdfix_macd'], df['ta_momentum_macdfix_signal'], df[
'ta_momentum_macdfix_hist'] = ta.MACDFIX(close, signalperiod=9)
df['ta_momentum_mfi'] = ta.MFI(high, low, close, volume, timeperiod=14)
df['ta_momentum_minus_di'] = ta.MINUS_DI(high,
low,
close,
timeperiod=14)
df['ta_momentum_minus_dm'] = ta.MINUS_DM(high, low, timeperiod=14)
df['ta_momentum_mom'] = ta.MOM(close, timeperiod=10)
df['ta_momentum_plus_di'] = ta.PLUS_DI(high, low, close, timeperiod=14)
df['ta_momentum_plus_dm'] = ta.PLUS_DM(high, low, timeperiod=14)
df['ta_momentum_ppo'] = ta.PPO(close,
fastperiod=12,
slowperiod=26,
matype=0)
df['ta_momentum_roc'] = ta.ROC(close, timeperiod=10)
df['ta_momentum_rocp'] = ta.ROCP(close, timeperiod=10)
df['ta_momentum_rocr'] = ta.ROCR(close, timeperiod=10)
df['ta_momentum_rocr100'] = ta.ROCR100(close, timeperiod=10)
df['ta_momentum_rsi'] = ta.RSI(close, timeperiod=14)
df['ta_momentum_slowk'], df['ta_momentum_slowd'] = ta.STOCH(
high,
low,
close,
fastk_period=5,
slowk_period=3,
slowk_matype=0,
slowd_period=3,
slowd_matype=0)
df['ta_momentum_fastk'], df['ta_momentum_fastd'] = ta.STOCHF(
high, low, close, fastk_period=5, fastd_period=3, fastd_matype=0)
df['ta_momentum_fastk'], df['ta_momentum_fastd'] = ta.STOCHRSI(
close,
timeperiod=14,
fastk_period=5,
fastd_period=3,
fastd_matype=0)
df['ta_momentum_trix'] = ta.TRIX(
close, timeperiod=12) # NOTE: Changed to avoid a lot of Nan values
df['ta_momentum_ultosc'] = ta.ULTOSC(high,
low,
close,
timeperiod1=7,
timeperiod2=14,
timeperiod3=28)
df['ta_momentum_willr'] = ta.WILLR(high, low, close, timeperiod=14)
if ta_settings['volume']:
df['ta_volume_ad'] = ta.AD(high, low, close, volume)
df['ta_volume_adosc'] = ta.ADOSC(high,
low,
close,
volume,
fastperiod=3,
slowperiod=10)
df['ta_volume_obv'] = ta.OBV(close, volume)
if ta_settings['volatility']:
df['ta_volatility_atr'] = ta.ATR(high, low, close, timeperiod=14)
df['ta_volatility_natr'] = ta.NATR(high, low, close, timeperiod=14)
df['ta_volatility_trange'] = ta.TRANGE(high, low, close)
if ta_settings['price']:
df['ta_price_avgprice'] = ta.AVGPRICE(open, high, low, close)
df['ta_price_medprice'] = ta.MEDPRICE(high, low)
df['ta_price_typprice'] = ta.TYPPRICE(high, low, close)
df['ta_price_wclprice'] = ta.WCLPRICE(high, low, close)
if ta_settings['cycle']:
df['ta_cycle_ht_dcperiod'] = ta.HT_DCPERIOD(close)
df['ta_cycle_ht_phasor_inphase'], df[
'ta_cycle_ht_phasor_quadrature'] = ta.HT_PHASOR(close)
df['ta_cycle_ht_trendmode'] = ta.HT_TRENDMODE(close)
# NOTE: Commented to avoid a lot of Nan values
# df['ta_cycle_ht_dcphase'] = ta.HT_DCPHASE(close)
# df['ta_cycle_ht_sine_sine'], df['ta_cycle_ht_sine_leadsine'] = ta.HT_SINE(close)
if ta_settings['pattern']:
df['ta_pattern_cdl2crows'] = ta.CDL2CROWS(open, high, low, close)
df['ta_pattern_cdl3blackrows'] = ta.CDL3BLACKCROWS(
open, high, low, close)
df['ta_pattern_cdl3inside'] = ta.CDL3INSIDE(open, high, low, close)
df['ta_pattern_cdl3linestrike'] = ta.CDL3LINESTRIKE(
open, high, low, close)
df['ta_pattern_cdl3outside'] = ta.CDL3OUTSIDE(open, high, low, close)
df['ta_pattern_cdl3starsinsouth'] = ta.CDL3STARSINSOUTH(
open, high, low, close)
df['ta_pattern_cdl3whitesoldiers'] = ta.CDL3WHITESOLDIERS(
open, high, low, close)
df['ta_pattern_cdlabandonedbaby'] = ta.CDLABANDONEDBABY(open,
high,
low,
close,
penetration=0)
df['ta_pattern_cdladvanceblock'] = ta.CDLADVANCEBLOCK(
open, high, low, close)
df['ta_pattern_cdlbelthold'] = ta.CDLBELTHOLD(open, high, low, close)
df['ta_pattern_cdlbreakaway'] = ta.CDLBREAKAWAY(open, high, low, close)
df['ta_pattern_cdlclosingmarubozu'] = ta.CDLCLOSINGMARUBOZU(
open, high, low, close)
df['ta_pattern_cdlconcealbabyswall'] = ta.CDLCONCEALBABYSWALL(
open, high, low, close)
df['ta_pattern_cdlcounterattack'] = ta.CDLCOUNTERATTACK(
open, high, low, close)
df['ta_pattern_cdldarkcloudcover'] = ta.CDLDARKCLOUDCOVER(
open, high, low, close, penetration=0)
df['ta_pattern_cdldoji'] = ta.CDLDOJI(open, high, low, close)
df['ta_pattern_cdldojistar'] = ta.CDLDOJISTAR(open, high, low, close)
df['ta_pattern_cdldragonflydoji'] = ta.CDLDRAGONFLYDOJI(
open, high, low, close)
df['ta_pattern_cdlengulfing'] = ta.CDLENGULFING(open, high, low, close)
df['ta_pattern_cdleveningdojistar'] = ta.CDLEVENINGDOJISTAR(
open, high, low, close, penetration=0)
df['ta_pattern_cdleveningstar'] = ta.CDLEVENINGSTAR(open,
high,
low,
close,
penetration=0)
df['ta_pattern_cdlgapsidesidewhite'] = ta.CDLGAPSIDESIDEWHITE(
open, high, low, close)
df['ta_pattern_cdlgravestonedoji'] = ta.CDLGRAVESTONEDOJI(
open, high, low, close)
df['ta_pattern_cdlhammer'] = ta.CDLHAMMER(open, high, low, close)
df['ta_pattern_cdlhangingman'] = ta.CDLHANGINGMAN(
open, high, low, close)
df['ta_pattern_cdlharami'] = ta.CDLHARAMI(open, high, low, close)
df['ta_pattern_cdlharamicross'] = ta.CDLHARAMICROSS(
open, high, low, close)
df['ta_pattern_cdlhighwave'] = ta.CDLHIGHWAVE(open, high, low, close)
df['ta_pattern_cdlhikkake'] = ta.CDLHIKKAKE(open, high, low, close)
df['ta_pattern_cdlhikkakemod'] = ta.CDLHIKKAKEMOD(
open, high, low, close)
df['ta_pattern_cdlhomingpigeon'] = ta.CDLHOMINGPIGEON(
open, high, low, close)
df['ta_pattern_cdlidentical3crows'] = ta.CDLIDENTICAL3CROWS(
open, high, low, close)
df['ta_pattern_cdlinneck'] = ta.CDLINNECK(open, high, low, close)
df['ta_pattern_cdlinvertedhammer'] = ta.CDLINVERTEDHAMMER(
open, high, low, close)
df['ta_pattern_cdlkicking'] = ta.CDLKICKING(open, high, low, close)
df['ta_pattern_cdlkickingbylength'] = ta.CDLKICKINGBYLENGTH(
open, high, low, close)
df['ta_pattern_cdlladderbottom'] = ta.CDLLADDERBOTTOM(
open, high, low, close)
df['ta_pattern_cdllongleggeddoji'] = ta.CDLLONGLEGGEDDOJI(
open, high, low, close)
df['ta_pattern_cdllongline'] = ta.CDLLONGLINE(open, high, low, close)
df['ta_pattern_cdlmarubozu'] = ta.CDLMARUBOZU(open, high, low, close)
df['ta_pattern_cdlmatchinglow'] = ta.CDLMATCHINGLOW(
open, high, low, close)
df['ta_pattern_cdlmathold'] = ta.CDLMATHOLD(open,
high,
low,
close,
penetration=0)
df['ta_pattern_cdlmorningdojistar'] = ta.CDLMORNINGDOJISTAR(
open, high, low, close, penetration=0)
df['ta_pattern_cdlmorningstar'] = ta.CDLMORNINGSTAR(open,
high,
low,
close,
penetration=0)
df['ta_pattern_cdllonneck'] = ta.CDLONNECK(open, high, low, close)
df['ta_pattern_cdlpiercing'] = ta.CDLPIERCING(open, high, low, close)
df['ta_pattern_cdlrickshawman'] = ta.CDLRICKSHAWMAN(
open, high, low, close)
df['ta_pattern_cdlrisefall3methods'] = ta.CDLRISEFALL3METHODS(
open, high, low, close)
df['ta_pattern_cdlseparatinglines'] = ta.CDLSEPARATINGLINES(
open, high, low, close)
df['ta_pattern_cdlshootingstar'] = ta.CDLSHOOTINGSTAR(
open, high, low, close)
df['ta_pattern_cdlshortline'] = ta.CDLSHORTLINE(open, high, low, close)
df['ta_pattern_cdlspinningtop'] = ta.CDLSPINNINGTOP(
open, high, low, close)
df['ta_pattern_cdlstalledpattern'] = ta.CDLSTALLEDPATTERN(
open, high, low, close)
df['ta_pattern_cdlsticksandwich'] = ta.CDLSTICKSANDWICH(
open, high, low, close)
df['ta_pattern_cdltakuri'] = ta.CDLTAKURI(open, high, low, close)
df['ta_pattern_cdltasukigap'] = ta.CDLTASUKIGAP(open, high, low, close)
df['ta_pattern_cdlthrusting'] = ta.CDLTHRUSTING(open, high, low, close)
df['ta_pattern_cdltristar'] = ta.CDLTRISTAR(open, high, low, close)
df['ta_pattern_cdlunique3river'] = ta.CDLUNIQUE3RIVER(
open, high, low, close)
df['ta_pattern_cdlupsidegap2crows'] = ta.CDLUPSIDEGAP2CROWS(
open, high, low, close)
df['ta_pattern_cdlxsidegap3methods'] = ta.CDLXSIDEGAP3METHODS(
open, high, low, close)
if ta_settings['statistic']:
df['ta_statistic_beta'] = ta.BETA(high, low, timeperiod=5)
df['ta_statistic_correl'] = ta.CORREL(high, low, timeperiod=30)
df['ta_statistic_linearreg'] = ta.LINEARREG(close, timeperiod=14)
df['ta_statistic_linearreg_angle'] = ta.LINEARREG_ANGLE(close,
timeperiod=14)
df['ta_statistic_linearreg_intercept'] = ta.LINEARREG_INTERCEPT(
close, timeperiod=14)
df['ta_statistic_linearreg_slope'] = ta.LINEARREG_SLOPE(close,
timeperiod=14)
df['ta_statistic_stddev'] = ta.STDDEV(close, timeperiod=5, nbdev=1)
df['ta_statistic_tsf'] = ta.TSF(close, timeperiod=14)
df['ta_statistic_var'] = ta.VAR(close, timeperiod=5, nbdev=1)
if ta_settings['math_transforms']:
df['ta_math_transforms_atan'] = ta.ATAN(close)
df['ta_math_transforms_ceil'] = ta.CEIL(close)
df['ta_math_transforms_cos'] = ta.COS(close)
df['ta_math_transforms_floor'] = ta.FLOOR(close)
df['ta_math_transforms_ln'] = ta.LN(close)
df['ta_math_transforms_log10'] = ta.LOG10(close)
df['ta_math_transforms_sin'] = ta.SIN(close)
df['ta_math_transforms_sqrt'] = ta.SQRT(close)
df['ta_math_transforms_tan'] = ta.TAN(close)
if ta_settings['math_operators']:
df['ta_math_operators_add'] = ta.ADD(high, low)
df['ta_math_operators_div'] = ta.DIV(high, low)
df['ta_math_operators_min'], df['ta_math_operators_max'] = ta.MINMAX(
close, timeperiod=30)
df['ta_math_operators_minidx'], df[
'ta_math_operators_maxidx'] = ta.MINMAXINDEX(close, timeperiod=30)
df['ta_math_operators_mult'] = ta.MULT(high, low)
df['ta_math_operators_sub'] = ta.SUB(high, low)
df['ta_math_operators_sum'] = ta.SUM(close, timeperiod=30)
return df
def LINEARREG_ANGLE(Series, timeperiod=14):
res = talib.LINEARREG_ANGLE(Series.values, timeperiod)
return pd.Series(res, index=Series.index)
def LINEARREG_ANGLE(raw_df, timeperiod=14):
# Linear Regression Angle
# extract necessary data from raw dataframe (close)
return ta.LINEARREG_ANGLE(raw_df.Close.values, timeperiod)
def get_talib_stock_daily(
stock_code,
s,
e,
append_ori_close=False,
norms=['volume', 'amount', 'ht_dcphase', 'obv', 'adosc', 'ad', 'cci']):
"""获取经过talib处理后的股票日线数据"""
stock_data = QA.QA_fetch_stock_day_adv(stock_code, s, e)
stock_df = stock_data.to_qfq().data
if append_ori_close:
stock_df['o_close'] = stock_data.data['close']
# stock_df['high_qfq'] = stock_data.to_qfq().data['high']
# stock_df['low_hfq'] = stock_data.to_hfq().data['low']
close = np.array(stock_df['close'])
high = np.array(stock_df['high'])
low = np.array(stock_df['low'])
_open = np.array(stock_df['open'])
_volume = np.array(stock_df['volume'])
stock_df['dema'] = talib.DEMA(close)
stock_df['ema'] = talib.EMA(close)
stock_df['ht_tradeline'] = talib.HT_TRENDLINE(close)
stock_df['kama'] = talib.KAMA(close)
stock_df['ma'] = talib.MA(close)
stock_df['mama'], stock_df['fama'] = talib.MAMA(close)
# MAVP
stock_df['midpoint'] = talib.MIDPOINT(close)
stock_df['midprice'] = talib.MIDPRICE(high, low)
stock_df['sar'] = talib.SAR(high, low)
stock_df['sarext'] = talib.SAREXT(high, low)
stock_df['sma'] = talib.SMA(close)
stock_df['t3'] = talib.T3(close)
stock_df['tema'] = talib.TEMA(close)
stock_df['trima'] = talib.TRIMA(close)
stock_df['wma'] = talib.WMA(close)
stock_df['adx'] = talib.ADX(high, low, close)
stock_df['adxr'] = talib.ADXR(high, low, close)
stock_df['apo'] = talib.APO(close)
stock_df['aroondown'], stock_df['aroonup'] = talib.AROON(high, low)
stock_df['aroonosc'] = talib.AROONOSC(high, low)
stock_df['bop'] = talib.BOP(_open, high, low, close)
stock_df['cci'] = talib.CCI(high, low, close)
stock_df['cmo'] = talib.CMO(close)
stock_df['dx'] = talib.DX(high, low, close)
# MACD
stock_df['macd'], stock_df['macdsignal'], stock_df[
'macdhist'] = talib.MACDEXT(close)
# MACDFIX
stock_df['mfi'] = talib.MFI(high, low, close, _volume)
stock_df['minus_di'] = talib.MINUS_DI(high, low, close)
stock_df['minus_dm'] = talib.MINUS_DM(high, low)
stock_df['mom'] = talib.MOM(close)
stock_df['plus_di'] = talib.PLUS_DI(high, low, close)
stock_df['plus_dm'] = talib.PLUS_DM(high, low)
stock_df['ppo'] = talib.PPO(close)
stock_df['roc'] = talib.ROC(close)
stock_df['rocp'] = talib.ROCP(close)
stock_df['rocr'] = talib.ROCR(close)
stock_df['rocr100'] = talib.ROCR100(close)
stock_df['rsi'] = talib.RSI(close)
stock_df['slowk'], stock_df['slowd'] = talib.STOCH(high, low, close)
stock_df['fastk'], stock_df['fastd'] = talib.STOCHF(high, low, close)
# STOCHRSI - Stochastic Relative Strength Index
stock_df['trix'] = talib.TRIX(close)
stock_df['ultosc'] = talib.ULTOSC(high, low, close)
stock_df['willr'] = talib.WILLR(high, low, close)
stock_df['ad'] = talib.AD(high, low, close, _volume)
stock_df['adosc'] = talib.ADOSC(high, low, close, _volume)
stock_df['obv'] = talib.OBV(close, _volume)
stock_df['ht_dcperiod'] = talib.HT_DCPERIOD(close)
stock_df['ht_dcphase'] = talib.HT_DCPHASE(close)
stock_df['inphase'], stock_df['quadrature'] = talib.HT_PHASOR(close)
stock_df['sine'], stock_df['leadsine'] = talib.HT_PHASOR(close)
stock_df['ht_trendmode'] = talib.HT_TRENDMODE(close)
stock_df['avgprice'] = talib.AVGPRICE(_open, high, low, close)
stock_df['medprice'] = talib.MEDPRICE(high, low)
stock_df['typprice'] = talib.TYPPRICE(high, low, close)
stock_df['wclprice'] = talib.WCLPRICE(high, low, close)
stock_df['atr'] = talib.ATR(high, low, close)
stock_df['natr'] = talib.NATR(high, low, close)
stock_df['trange'] = talib.TRANGE(high, low, close)
stock_df['beta'] = talib.BETA(high, low)
stock_df['correl'] = talib.CORREL(high, low)
stock_df['linearreg'] = talib.LINEARREG(close)
stock_df['linearreg_angle'] = talib.LINEARREG_ANGLE(close)
stock_df['linearreg_intercept'] = talib.LINEARREG_INTERCEPT(close)
stock_df['linearreg_slope'] = talib.LINEARREG_SLOPE(close)
stock_df['stddev'] = talib.STDDEV(close)
stock_df['tsf'] = talib.TSF(close)
stock_df['var'] = talib.VAR(close)
stock_df = stock_df.reset_index().set_index('date')
if norms:
x = stock_df[norms].values # returns a numpy array
x_scaled = MinMaxScaler().fit_transform(x)
stock_df = stock_df.drop(columns=norms).join(
pd.DataFrame(x_scaled, columns=norms, index=stock_df.index))
# stock_df = stock_df.drop(columns=['code', 'open', 'high', 'low'])
stock_df = stock_df.dropna()
stock_df = stock_df.drop(columns=['code'])
return stock_df
def __prepare_one_hold(df, _back_hours, _hold_hour, diff_d=[0.3, 0.5]):
""" 为一个币种一个持币周期添加所有回溯周期的因子数据，并添加该持币周期的所有offset，返回一个 DataFrame
:param _pkl_file: 整理好基础数据的一个币种的 pkl 文件路径
:param _back_hours: 回溯周期列表 [3, 4, 6, 8, 12, 24, 48, 60, 72, 96] 关联因子周期
skew偏度rolling最小周期为3才有数据 所以最小回溯周期设为3
:param _hold_hour: 持币周期 '2H', '3H', '4H', '6H', '8H', '12H', '24H', '36H', '48H', '60H', '72H'
上述周期中之一， 关联 添加 offset 标签列
如果持币周为 2H offset 为 0, 1; 如果持币周为 3H offset 为 0, 1, 2;
:return: """
df['涨跌幅'] = df['close'].pct_change() # 计算涨跌幅
df['下个周期_avg_price'] = df['avg_price'].shift(-1) # 计算下根K线开盘买入涨跌幅
df.loc[df['volume'] == 0, '是否交易'] = 0 # 找出不交易的周期
df['是否交易'].fillna(value=1, inplace=True)
""" ******************** 以下是需要修改的代码 ******************** """
# =====计算各项选币指标
extra_agg_dict = dict()
# =====技术指标
# --- KDJ ---
for n in _back_hours:
# 正常K线数据 计算 KDJ
low_list = df['low'].rolling(
n, min_periods=1).min() # 过去n(含当前行)行数据 最低价的最小值
high_list = df['high'].rolling(
n, min_periods=1).max() # 过去n(含当前行)行数据 最高价的最大值
rsv = (df['close'] - low_list) / (high_list -
low_list) * 100 # 未成熟随机指标值
df[f'K_bh_{n}'] = rsv.ewm(com=2).mean().shift(1) # K
extra_agg_dict[f'K_bh_{n}'] = 'first'
df[f'D_bh_{n}'] = df[f'K_bh_{n}'].ewm(com=2).mean() # D
extra_agg_dict[f'D_bh_{n}'] = 'first'
df[f'J_bh_{n}'] = 3 * df[f'K_bh_{n}'] - 2 * df[f'D_bh_{n}'] # J
extra_agg_dict[f'J_bh_{n}'] = 'first'
# 差分
# 使用差分后的K线数据 计算 KDJ --- 标准差变大，数据更不稳定，放弃
# 用计算后的KDJ指标，再差分 --- 标准差变小，数据更稳定，采纳
for _ind in ['K', 'D', 'J']:
__add_diff(_df=df,
_d_list=diff_d,
_name=f'{_ind}_bh_{n}',
_agg_dict=extra_agg_dict,
_agg_type='first')
# --- RSI --- 在期货市场很有效
close_dif = df['close'].diff()
df['up'] = np.where(close_dif > 0, close_dif, 0)
df['down'] = np.where(close_dif < 0, abs(close_dif), 0)
for n in _back_hours:
a = df['up'].rolling(n).sum()
b = df['down'].rolling(n).sum()
df[f'RSI_bh_{n}'] = (a / (a + b)).shift(1) # RSI
extra_agg_dict[f'RSI_bh_{n}'] = 'first'
# 差分
# 用计算后的RSI指标，再差分 --- 标准差变小，数据更稳定，采纳
__add_diff(_df=df,
_d_list=diff_d,
_name=f'RSI_bh_{n}',
_agg_dict=extra_agg_dict,
_agg_type='first')
del df['up'], df['down'] # 删除过程数据
# ===常见变量
# --- 均价 --- 对应低价股策略(预计没什么用)
# 策略改进思路：以下所有用到收盘价的因子，都可尝试使用均价代替
for n in _back_hours:
df[f'均价_bh_{n}'] = (df['quote_volume'].rolling(n).sum() /
df['volume'].rolling(n).sum()).shift(1)
extra_agg_dict[f'均价_bh_{n}'] = 'first'
# 差分
__add_diff(_df=df,
_d_list=diff_d,
_name=f'均价_bh_{n}',
_agg_dict=extra_agg_dict,
_agg_type='first')
# --- 涨跌幅 ---
for n in _back_hours:
df[f'涨跌幅_bh_{n}'] = df['close'].pct_change(n).shift(1)
extra_agg_dict[f'涨跌幅_bh_{n}'] = 'first'
# 差分
__add_diff(_df=df,
_d_list=diff_d,
_name=f'涨跌幅_bh_{n}',
_agg_dict=extra_agg_dict,
_agg_type='first')
# --- bias --- 涨跌幅更好的表达方式 bias 币价偏离均线的比例。
for n in _back_hours:
ma = df['close'].rolling(n, min_periods=1).mean()
df[f'bias_bh_{n}'] = (df['close'] / ma - 1).shift(1)
extra_agg_dict[f'bias_bh_{n}'] = 'first'
# 差分
__add_diff(_df=df,
_d_list=diff_d,
_name=f'bias_bh_{n}',
_agg_dict=extra_agg_dict,
_agg_type='first')
# --- 振幅 --- 最高价最低价
for n in _back_hours:
high = df['high'].rolling(n, min_periods=1).max()
low = df['low'].rolling(n, min_periods=1).min()
df[f'振幅_bh_{n}'] = (high / low - 1).shift(1)
extra_agg_dict[f'振幅_bh_{n}'] = 'first'
# 差分
__add_diff(_df=df,
_d_list=diff_d,
_name=f'振幅_bh_{n}',
_agg_dict=extra_agg_dict,
_agg_type='first')
# --- 振幅2 --- 收盘价、开盘价
high = df[['close', 'open']].max(axis=1)
low = df[['close', 'open']].min(axis=1)
for n in _back_hours:
high = high.rolling(n, min_periods=1).max()
low = low.rolling(n, min_periods=1).min()
df[f'振幅2_bh_{n}'] = (high / low - 1).shift(1)
extra_agg_dict[f'振幅2_bh_{n}'] = 'first'
# 差分
__add_diff(_df=df,
_d_list=diff_d,
_name=f'振幅2_bh_{n}',
_agg_dict=extra_agg_dict,
_agg_type='first')
# --- 涨跌幅std --- 振幅的另外一种形式
change = df['close'].pct_change()
for n in _back_hours:
df[f'涨跌幅std_bh_{n}'] = change.rolling(n).std().shift(1)
extra_agg_dict[f'涨跌幅std_bh_{n}'] = 'first'
# 差分
__add_diff(_df=df,
_d_list=diff_d,
_name=f'涨跌幅std_bh_{n}',
_agg_dict=extra_agg_dict,
_agg_type='first')
# --- 涨跌幅skew --- 在商品期货市场有效
# skew偏度rolling最小周期为3才有数据
for n in _back_hours:
df[f'涨跌幅skew_bh_{n}'] = change.rolling(n).skew().shift(1)
extra_agg_dict[f'涨跌幅skew_bh_{n}'] = 'first'
# 差分
__add_diff(_df=df,
_d_list=diff_d,
_name=f'涨跌幅skew_bh_{n}',
_agg_dict=extra_agg_dict,
_agg_type='first')
# --- 成交额 --- 对应小市值概念
for n in _back_hours:
df[f'成交额_bh_{n}'] = df['quote_volume'].rolling(
n, min_periods=1).sum().shift(1)
extra_agg_dict[f'成交额_bh_{n}'] = 'first'
# 差分
__add_diff(_df=df,
_d_list=diff_d,
_name=f'成交额_bh_{n}',
_agg_dict=extra_agg_dict,
_agg_type='first')
# --- 成交额std --- 191选股因子中最有效的因子
for n in _back_hours:
df[f'成交额std_bh_{n}'] = df['quote_volume'].rolling(
n, min_periods=2).std().shift(1)
extra_agg_dict[f'成交额std_bh_{n}'] = 'first'
# 差分
__add_diff(_df=df,
_d_list=diff_d,
_name=f'成交额std_bh_{n}',
_agg_dict=extra_agg_dict,
_agg_type='first')
# --- 资金流入比例 --- 币安独有的数据
for n in _back_hours:
volume = df['quote_volume'].rolling(n, min_periods=1).sum()
buy_volume = df['taker_buy_quote_asset_volume'].rolling(
n, min_periods=1).sum()
df[f'资金流入比例_bh_{n}'] = (buy_volume / volume).shift(1)
extra_agg_dict[f'资金流入比例_bh_{n}'] = 'first'
# 差分
__add_diff(_df=df,
_d_list=diff_d,
_name=f'资金流入比例_bh_{n}',
_agg_dict=extra_agg_dict,
_agg_type='first')
# --- 量比 ---
for n in _back_hours:
df[f'量比_bh_{n}'] = (
df['quote_volume'] /
df['quote_volume'].rolling(n, min_periods=1).mean()).shift(1)
extra_agg_dict[f'量比_bh_{n}'] = 'first'
# 差分
__add_diff(_df=df,
_d_list=diff_d,
_name=f'量比_bh_{n}',
_agg_dict=extra_agg_dict,
_agg_type='first')
# --- 成交笔数 ---
for n in _back_hours:
df[f'成交笔数_bh_{n}'] = df['trade_num'].rolling(
n, min_periods=1).sum().shift(1)
extra_agg_dict[f'成交笔数_bh_{n}'] = 'first'
# 差分
__add_diff(_df=df,
_d_list=diff_d,
_name=f'成交笔数_bh_{n}',
_agg_dict=extra_agg_dict,
_agg_type='first')
# --- 量价相关系数 --- 量价相关选股策略
for n in _back_hours:
df[f'量价相关系数_bh_{n}'] = df['close'].rolling(n).corr(
df['quote_volume']).shift(1)
extra_agg_dict[f'量价相关系数_bh_{n}'] = 'first'
# 差分
__add_diff(_df=df,
_d_list=diff_d,
_name=f'量价相关系数_bh_{n}',
_agg_dict=extra_agg_dict,
_agg_type='first')
# --- Angle ---
for n in _back_hours:
column_name = f'Angle_bh_{n}'
ma = df['close'].rolling(window=n, min_periods=1).mean()
df[column_name] = ta.LINEARREG_ANGLE(ma, n)
df[column_name] = df[column_name].shift(1)
extra_agg_dict[column_name] = 'first'
# 差分
__add_diff(_df=df,
_d_list=diff_d,
_name=column_name,
_agg_dict=extra_agg_dict,
_agg_type='first')
# ---- GapTrue ----
for n in _back_hours:
ma = df['close'].rolling(window=n, min_periods=1).mean()
wma = ta.WMA(df['close'], n)
gap = wma - ma
column_name = f'GapTrue_bh_{n}'
df[column_name] = gap / abs(gap).rolling(window=n).sum()
df[column_name] = df[column_name].shift(1)
extra_agg_dict[column_name] = 'first'
# 差分
__add_diff(_df=df,
_d_list=diff_d,
_name=column_name,
_agg_dict=extra_agg_dict,
_agg_type='first')
# ---- 癞子 ----
for n in _back_hours:
diff = df['close'] / df['close'].shift(1) - 1
column_name = f'癞子_bh_{n}'
df[column_name] = diff / abs(diff).rolling(window=n).sum()
df[column_name] = df[column_name].shift(1)
extra_agg_dict[column_name] = 'first'
# 差分
__add_diff(_df=df,
_d_list=diff_d,
_name=column_name,
_agg_dict=extra_agg_dict,
_agg_type='first')
# ---- CCI ----
for n in _back_hours:
oma = ta.WMA(df.open, n)
hma = ta.WMA(df.high, n)
lma = ta.WMA(df.low, n)
cma = ta.WMA(df.close, n)
tp = (hma + lma + cma + oma) / 4
ma = ta.WMA(tp, n)
md = ta.WMA(abs(cma - ma), n)
column_name = f'CCI_bh_{n}'
df[column_name] = (tp - ma) / md
df[column_name] = df[column_name].shift(1)
extra_agg_dict[column_name] = 'first'
# 差分
__add_diff(_df=df,
_d_list=diff_d,
_name=column_name,
_agg_dict=extra_agg_dict,
_agg_type='first')
""" ******************** 以上是需要修改的代码 ******************** """
# ===将数据转化为需要的周期
# 在数据最前面，增加一行数据，这是为了在对>24h的周期进行resample时，保持数据的一致性。
df = df.loc[0:0, :].append(df, ignore_index=True)
df.loc[0, 'candle_begin_time'] = dfr.convert_string_to_local_date(
'2020-01-01 00:00:00').replace(tzinfo=pytz.utc)
# 转换周期
df['周期开始时间'] = df['candle_begin_time']
df.set_index('candle_begin_time', inplace=True)
# 必备字段
agg_dict = {
'symbol': 'first',
'周期开始时间': 'first',
'open': 'first',
'avg_price': 'first',
'close': 'last',
'下个周期_avg_price': 'last',
'volume': 'sum',
}
agg_dict = dict(agg_dict, **extra_agg_dict) # 需要保留的列 必备字段 + 因子字段
# 不同的offset，进行resample
# 不管何种持币周期，将所有 offset 合并为一个 DataFrame后 一天中的数据行数都是相同的
# 一天的数据行数为 当天的币种数*24,
# 例如在2020-12-15有77个币在交易，不管持币周期是2H还是3H，合并后的DataFrame数据行数都是1848(24*77) 因为
# 对于持币周期为2H来说，offset0 offset1 共计两批换仓点
# offset0 换仓点在 0，2，4，6，8，10，12，14，16，18，20，22
# offset1 换仓点在 1，3，5，7，9，11，13，15，17，19，21，23
# 所以在一天中的每个整点都有所有交易对的换仓发生
# 对于持币周期为3H来说，offset0 offset1 offset2 共计三批换仓点
# offset0 换仓点在 0，3，6，9，12，15，18，21
# offset1 换仓点在 1，4，7，10，13，16，19，22
# offset2 换仓点在 2，5，8，11，14，17，20，23
# 所以在一天中的每个整点都有所有交易对的换仓发生
# 对于持币周期大于一天的来说，持币周期内相当于一天，例如48H，还是有48个offset 共计48批换仓点
# 所以在固定长度日期范围内，所有持币周期都将有相同的数据行数
# 所以不管何种持币周期，生成的合并 pkl 大小基本相同
period_df_list = []
for offset in range(int(_hold_hour[:-1])):
# 转换持币周期的 df 其列为 必备字段 + 因子字段
# 开始时间为 2010-01-01 00:00:00 在有交易数据的之前 除candle_begin_time列皆为 NaN
# 在24h之内 base 和 offset 一样
# period_df = df.resample(_hold_hour, offset=f'{offset}h').agg(agg_dict)
period_df = df.resample(_hold_hour, base=offset).agg(agg_dict)
# 上一行代码对数据转换完周期后，刚开始会有大量的空数据，没有必要对这些数据进行删除，因为后面会删除18年之前的数据。
# 添加 offset 标签列
# 如果持币周为 2 offset 为 0, 1; 如果持币周为 3 offset 为 0, 1, 2;
period_df['offset'] = offset
# 原版自适应布林通道
n = 34
period_df['close_shift'] = period_df['close'].shift(1)
period_df['median'] = period_df['close_shift'].rolling(window=n).mean()
period_df['std'] = period_df['close_shift'].rolling(
n, min_periods=1).std(ddof=0) # ddof代表标准差自由度
period_df['z_score'] = abs(period_df['close_shift'] -
period_df['median']) / period_df['std']
period_df['up'] = period_df['z_score'].rolling(
window=n, min_periods=1).max().shift(1)
period_df['dn'] = period_df['z_score'].rolling(
window=n, min_periods=1).min().shift(1)
period_df[
'upper'] = period_df['median'] + period_df['std'] * period_df['up']
period_df[
'lower'] = period_df['median'] - period_df['std'] * period_df['up']
period_df['condition_long'] = period_df['close_shift'] >= period_df[
'lower'] # 破下轨，不做多
period_df['condition_short'] = period_df['close_shift'] <= period_df[
'upper'] # 破上轨，不做空
period_df.reset_index(inplace=True)
# 合并数据
period_df_list.append(period_df)
# 将不同offset的数据，合并到一张表
period_df = pd.concat(period_df_list, ignore_index=True)
period_df.reset_index(inplace=True)
# 删除一些数据
period_df = period_df.iloc[24:] # 刚开始交易前24个周期删除
period_df = period_df[period_df['candle_begin_time'] >= pd.to_datetime(
'2020-09-01').tz_localize(pytz.utc)]
period_df.reset_index(drop=True, inplace=True)
del period_df['index']
return period_df
def _extract_feature(candle, params, candle_type, target_dt):
'''
前に余分に必要なデータ量: {(stockf_fastk_period_l + stockf_fastk_period_l) * 最大分足 (min)} + window_size
= (12 + 12) * 5 + 5 = 125 (min)
'''
o = candle.open
h = candle.high
l = candle.low
c = candle.close
v = candle.volume
# OHLCV
features = pd.DataFrame()
features['open'] = o
features['high'] = h
features['low'] = l
features['close'] = c
features['volume'] = v
####################################
#
# Momentum Indicator Functions
#
####################################
# ADX = SUM((+DI - (-DI)) / (+DI + (-DI)), N) / N
# N — 計算期間
# SUM (..., N) — N期間の合計
# +DI — プラスの価格変動の値（positive directional index)
# -DI — マイナスの価格変動の値（negative directional index）
# rsi_timeperiod_l=30の場合、30分足で、(30 * 30 / 60(min)) = 15時間必要
features['adx_s'] = ta.ADX(h, l, c, timeperiod=params['adx_timeperiod_s'])
features['adx_m'] = ta.ADX(h, l, c, timeperiod=params['adx_timeperiod_m'])
features['adx_l'] = ta.ADX(h, l, c, timeperiod=params['adx_timeperiod_l'])
features['adxr_s'] = ta.ADXR(h, l, c, timeperiod=params['adxr_timeperiod_s'])
features['adxr_m'] = ta.ADXR(h, l, c, timeperiod=params['adxr_timeperiod_m'])
features['adxr_l'] = ta.ADXR(h, l, c, timeperiod=params['adxr_timeperiod_l'])
# APO = Shorter Period EMA – Longer Period EMA
features['apo_s'] = ta.APO(c, fastperiod=params['apo_fastperiod_s'], slowperiod=params['apo_slowperiod_s'], matype=ta.MA_Type.EMA)
features['apo_m'] = ta.APO(c, fastperiod=params['apo_fastperiod_m'], slowperiod=params['apo_slowperiod_m'], matype=ta.MA_Type.EMA)
# AroonUp = (N － 過去N日間の最高値からの経過期間) ÷ N × 100
# AroonDown = (N － 過去N日間の最安値からの経過期間) ÷ N × 100
# aroon_timeperiod_l=30の場合、30分足で、(30 * 30 / 60(min)) = 15時間必要
#features['aroondown_s'], features['aroonup_s'] = ta.AROON(h, l, timeperiod=params['aroon_timeperiod_s'])
#features['aroondown_m'], features['aroonup_m'] = ta.AROON(h, l, timeperiod=params['aroon_timeperiod_m'])
#features['aroondown_l'], features['aroonup_l'] = ta.AROON(h, l, timeperiod=params['aroon_timeperiod_l'])
# Aronnオシレーター = AroonUp － AroonDown
# aroonosc_timeperiod_l=30の場合、30分足で、(30 * 30 / 60(min)) = 15時間必要
features['aroonosc_s'] = ta.AROONOSC(h, l, timeperiod=params['aroonosc_timeperiod_s'])
features['aroonosc_m'] = ta.AROONOSC(h, l, timeperiod=params['aroonosc_timeperiod_m'])
features['aroonosc_l'] = ta.AROONOSC(h, l, timeperiod=params['aroonosc_timeperiod_l'])
# BOP = (close - open) / (high - low)
features['bop'] = ta.BOP(o, h, l, c)
# CCI = (TP - MA) / (0.015 * MD)
# TP: (高値+安値+終値) / 3
# MA: TPの移動平均
# MD: 平均偏差 = ((MA - TP1) + (MA - TP2) + ...) / N
features['cci_s'] = ta.CCI(h, l, c, timeperiod=params['cci_timeperiod_s'])
features['cci_m'] = ta.CCI(h, l, c, timeperiod=params['cci_timeperiod_m'])
features['cci_l'] = ta.CCI(h, l, c, timeperiod=params['cci_timeperiod_l'])
# CMO - Chande Momentum Oscillator
#features['cmo_s'] = ta.CMO(c, timeperiod=params['cmo_timeperiod_s'])
#features['cmo_m'] = ta.CMO(c, timeperiod=params['cmo_timeperiod_m'])
#features['cmo_l'] = ta.CMO(c, timeperiod=params['cmo_timeperiod_l'])
# DX - Directional Movement Index
features['dx_s'] = ta.DX(h, l, c, timeperiod=params['dx_timeperiod_s'])
features['dx_m'] = ta.DX(h, l, c, timeperiod=params['dx_timeperiod_m'])
features['dx_l'] = ta.DX(h, l, c, timeperiod=params['dx_timeperiod_l'])
# MACD＝基準線－相対線
# 基準線（EMA）：過去12日（週・月）間の終値指数平滑平均
# 相対線（EMA）：過去26日（週・月）間の終値指数平滑平均
# https://www.sevendata.co.jp/shihyou/technical/macd.html
# macd_slowperiod_m = 30 の場合30分足で（(30 + macd_signalperiod_m) * 30）/ 60 = 16.5時間必要（macd_signalperiod_m=3の時）
macd, macdsignal, macdhist = ta.MACDEXT(c, fastperiod=params['macd_fastperiod_s'],
slowperiod=params['macd_slowperiod_s'],
signalperiod=params['macd_signalperiod_s'],
fastmatype=ta.MA_Type.EMA, slowmatype=ta.MA_Type.EMA,
signalmatype=ta.MA_Type.EMA)
change_macd = calc_change(macd, macdsignal)
change_macd.index = macd.index
features['macd_s'] = macd
features['macdsignal_s'] = macdsignal
features['macdhist_s'] = macdhist
features['change_macd_s'] = change_macd
macd, macdsignal, macdhist = ta.MACDEXT(c, fastperiod=params['macd_fastperiod_m'],
slowperiod=params['macd_slowperiod_m'],
signalperiod=params['macd_signalperiod_m'],
fastmatype=ta.MA_Type.EMA, slowmatype=ta.MA_Type.EMA,
signalmatype=ta.MA_Type.EMA)
change_macd = calc_change(macd, macdsignal)
change_macd.index = macd.index
features['macd_m'] = macd
features['macdsignal_m'] = macdsignal
features['macdhist_m'] = macdhist
features['change_macd_m'] = change_macd
# MFI - Money Flow Index
features['mfi_s'] = ta.MFI(h, l, c, v, timeperiod=params['mfi_timeperiod_s'])
features['mfi_m'] = ta.MFI(h, l, c, v, timeperiod=params['mfi_timeperiod_m'])
features['mfi_l'] = ta.MFI(h, l, c, v, timeperiod=params['mfi_timeperiod_l'])
# MINUS_DI - Minus Directional Indicator
features['minus_di_s'] = ta.MINUS_DI(h, l, c, timeperiod=params['minus_di_timeperiod_s'])
features['minus_di_m'] = ta.MINUS_DI(h, l, c, timeperiod=params['minus_di_timeperiod_m'])
features['minus_di_l'] = ta.MINUS_DI(h, l, c, timeperiod=params['minus_di_timeperiod_l'])
# MINUS_DM - Minus Directional Movement
features['minus_dm_s'] = ta.MINUS_DM(h, l, timeperiod=params['minus_dm_timeperiod_s'])
features['minus_dm_m'] = ta.MINUS_DM(h, l, timeperiod=params['minus_dm_timeperiod_m'])
features['minus_dm_l'] = ta.MINUS_DM(h, l, timeperiod=params['minus_dm_timeperiod_l'])
# MOM - Momentum
features['mom_s'] = ta.MOM(c, timeperiod=params['mom_timeperiod_s'])
features['mom_m'] = ta.MOM(c, timeperiod=params['mom_timeperiod_m'])
features['mom_l'] = ta.MOM(c, timeperiod=params['mom_timeperiod_l'])
# PLUS_DI - Minus Directional Indicator
features['plus_di_s'] = ta.PLUS_DI(h, l, c, timeperiod=params['plus_di_timeperiod_s'])
features['plus_di_m'] = ta.PLUS_DI(h, l, c, timeperiod=params['plus_di_timeperiod_m'])
features['plus_di_l'] = ta.PLUS_DI(h, l, c, timeperiod=params['plus_di_timeperiod_l'])
# PLUS_DM - Minus Directional Movement
features['plus_dm_s'] = ta.PLUS_DM(h, l, timeperiod=params['plus_dm_timeperiod_s'])
features['plus_dm_m'] = ta.PLUS_DM(h, l, timeperiod=params['plus_dm_timeperiod_m'])
features['plus_dm_l'] = ta.PLUS_DM(h, l, timeperiod=params['plus_dm_timeperiod_l'])
# PPO - Percentage Price Oscillator
#features['ppo_s'] = ta.PPO(c, fastperiod=params['ppo_fastperiod_s'], slowperiod=params['ppo_slowperiod_s'], matype=ta.MA_Type.EMA)
#features['ppo_m'] = ta.PPO(c, fastperiod=params['ppo_fastperiod_m'], slowperiod=params['ppo_slowperiod_m'], matype=ta.MA_Type.EMA)
# ROC - Rate of change : ((price/prevPrice)-1)*100
features['roc_s'] = ta.ROC(c, timeperiod=params['roc_timeperiod_s'])
features['roc_m'] = ta.ROC(c, timeperiod=params['roc_timeperiod_m'])
features['roc_l'] = ta.ROC(c, timeperiod=params['roc_timeperiod_l'])
# ROCP = (price-prevPrice) / prevPrice
# http://www.tadoc.org/indicator/ROCP.htm
# rocp_timeperiod_l = 30 の場合、30分足で(30 * 30) / 60 = 15時間必要
rocp = ta.ROCP(c, timeperiod=params['rocp_timeperiod_s'])
change_rocp = calc_change(rocp, pd.Series(np.zeros(len(candle)), index=candle.index))
change_rocp.index = rocp.index
features['rocp_s'] = rocp
features['change_rocp_s'] = change_rocp
rocp = ta.ROCP(c, timeperiod=params['rocp_timeperiod_m'])
change_rocp = calc_change(rocp, pd.Series(np.zeros(len(candle)), index=candle.index))
change_rocp.index = rocp.index
features['rocp_m'] = rocp
features['change_rocp_m'] = change_rocp
rocp = ta.ROCP(c, timeperiod=params['rocp_timeperiod_l'])
change_rocp = calc_change(rocp, pd.Series(np.zeros(len(candle)), index=candle.index))
change_rocp.index = rocp.index
features['rocp_l'] = rocp
features['change_rocp_l'] = change_rocp
# ROCR - Rate of change ratio: (price/prevPrice)
features['rocr_s'] = ta.ROCR(c, timeperiod=params['rocr_timeperiod_s'])
features['rocr_m'] = ta.ROCR(c, timeperiod=params['rocr_timeperiod_m'])
features['rocr_l'] = ta.ROCR(c, timeperiod=params['rocr_timeperiod_l'])
# ROCR100 - Rate of change ratio 100 scale: (price/prevPrice)*100
features['rocr100_s'] = ta.ROCR100(c, timeperiod=params['rocr100_timeperiod_s'])
features['rocr100_m'] = ta.ROCR100(c, timeperiod=params['rocr100_timeperiod_m'])
features['rocr100_l'] = ta.ROCR100(c, timeperiod=params['rocr100_timeperiod_l'])
# RSI = (100 * a) / (a + b) (a: x日間の値上がり幅の合計, b: x日間の値下がり幅の合計)
# https://www.sevendata.co.jp/shihyou/technical/rsi.html
# rsi_timeperiod_l=30の場合、30分足で、(30 * 30 / 60(min)) = 15時間必要
#features['rsi_s'] = ta.RSI(c, timeperiod=params['rsi_timeperiod_s'])
#features['rsi_m'] = ta.RSI(c, timeperiod=params['rsi_timeperiod_m'])
#features['rsi_l'] = ta.RSI(c, timeperiod=params['rsi_timeperiod_l'])
# FASTK(KPeriod) = 100 * (Today's Close - LowestLow) / (HighestHigh - LowestLow)
# FASTD(FastDperiod) = MA Smoothed FASTK over FastDperiod
# http://www.tadoc.org/indicator/STOCHF.htm
# stockf_fastk_period_l=30の場合30分足で、(((30 + 30) * 30) / 60(min)) = 30時間必要 (LowestLowが移動平均の30分余分に必要なので60period余分に計算する)
fastk, fastd = ta.STOCHF(h, l, c, fastk_period=params['stockf_fastk_period_s'], fastd_period=params['stockf_fastd_period_s'], fastd_matype=ta.MA_Type.EMA)
change_stockf = calc_change(fastk, fastd)
change_stockf.index = fastk.index
features['fastk_s'] = fastk
features['fastd_s'] = fastd
features['fast_change_s'] = change_stockf
fastk, fastd = ta.STOCHF(h, l, c, fastk_period=params['stockf_fastk_period_m'], fastd_period=params['stockf_fastd_period_m'], fastd_matype=ta.MA_Type.EMA)
change_stockf = calc_change(fastk, fastd)
change_stockf.index = fastk.index
features['fastk_m'] = fastk
features['fastd_m'] = fastd
features['fast_change_m'] = change_stockf
fastk, fastd = ta.STOCHF(h, l, c, fastk_period=params['stockf_fastk_period_l'], fastd_period=params['stockf_fastk_period_l'], fastd_matype=ta.MA_Type.EMA)
change_stockf = calc_change(fastk, fastd)
change_stockf.index = fastk.index
features['fastk_l'] = fastk
features['fastd_l'] = fastd
features['fast_change_l'] = change_stockf
# TRIX - 1-day Rate-Of-Change (ROC) of a Triple Smooth EMA
features['trix_s'] = ta.TRIX(c, timeperiod=params['trix_timeperiod_s'])
features['trix_m'] = ta.TRIX(c, timeperiod=params['trix_timeperiod_m'])
features['trix_l'] = ta.TRIX(c, timeperiod=params['trix_timeperiod_l'])
# ULTOSC - Ultimate Oscillator
features['ultosc_s'] = ta.ULTOSC(h, l, c, timeperiod1=params['ultosc_timeperiod_s1'], timeperiod2=params['ultosc_timeperiod_s2'], timeperiod3=params['ultosc_timeperiod_s3'])
# WILLR = (当日終値 - N日間の最高値) / (N日間の最高値 - N日間の最安値)× 100
# https://inet-sec.co.jp/study/technical-manual/williamsr/
# willr_timeperiod_l=30の場合30分足で、(30 * 30 / 60) = 15時間必要
features['willr_s'] = ta.WILLR(h, l, c, timeperiod=params['willr_timeperiod_s'])
features['willr_m'] = ta.WILLR(h, l, c, timeperiod=params['willr_timeperiod_m'])
features['willr_l'] = ta.WILLR(h, l, c, timeperiod=params['willr_timeperiod_l'])
####################################
#
# Volume Indicator Functions
#
####################################
# Volume Indicator Functions
# slowperiod_adosc_s = 10の場合、30分足で(10 * 30) / 60 = 5時間必要
features['ad'] = ta.AD(h, l, c, v)
features['adosc_s'] = ta.ADOSC(h, l, c, v, fastperiod=params['fastperiod_adosc_s'], slowperiod=params['slowperiod_adosc_s'])
features['obv'] = ta.OBV(c, v)
####################################
#
# Volatility Indicator Functions
#
####################################
# ATR - Average True Range
features['atr_s'] = ta.ATR(h, l, c, timeperiod=params['atr_timeperiod_s'])
features['atr_m'] = ta.ATR(h, l, c, timeperiod=params['atr_timeperiod_m'])
features['atr_l'] = ta.ATR(h, l, c, timeperiod=params['atr_timeperiod_l'])
# NATR - Normalized Average True Range
#features['natr_s'] = ta.NATR(h, l, c, timeperiod=params['natr_timeperiod_s'])
#features['natr_m'] = ta.NATR(h, l, c, timeperiod=params['natr_timeperiod_m'])
#features['natr_l'] = ta.NATR(h, l, c, timeperiod=params['natr_timeperiod_l'])
# TRANGE - True Range
features['trange'] = ta.TRANGE(h, l, c)
####################################
#
# Price Transform Functions
#
####################################
features['avgprice'] = ta.AVGPRICE(o, h, l, c)
features['medprice'] = ta.MEDPRICE(h, l)
#features['typprice'] = ta.TYPPRICE(h, l, c)
#features['wclprice'] = ta.WCLPRICE(h, l, c)
####################################
#
# Cycle Indicator Functions
#
####################################
#features['ht_dcperiod'] = ta.HT_DCPERIOD(c)
#features['ht_dcphase'] = ta.HT_DCPHASE(c)
#features['inphase'], features['quadrature'] = ta.HT_PHASOR(c)
#features['sine'], features['leadsine'] = ta.HT_SINE(c)
features['integer'] = ta.HT_TRENDMODE(c)
####################################
#
# Statistic Functions
#
####################################
# BETA - Beta
features['beta_s'] = ta.BETA(h, l, timeperiod=params['beta_timeperiod_s'])
features['beta_m'] = ta.BETA(h, l, timeperiod=params['beta_timeperiod_m'])
features['beta_l'] = ta.BETA(h, l, timeperiod=params['beta_timeperiod_l'])
# CORREL - Pearson's Correlation Coefficient (r)
#features['correl_s'] = ta.CORREL(h, l, timeperiod=params['correl_timeperiod_s'])
#features['correl_m'] = ta.CORREL(h, l, timeperiod=params['correl_timeperiod_m'])
#features['correl_l'] = ta.CORREL(h, l, timeperiod=params['correl_timeperiod_l'])
# LINEARREG - Linear Regression
#features['linearreg_s'] = ta.LINEARREG(c, timeperiod=params['linearreg_timeperiod_s'])
#features['linearreg_m'] = ta.LINEARREG(c, timeperiod=params['linearreg_timeperiod_m'])
#features['linearreg_l'] = ta.LINEARREG(c, timeperiod=params['linearreg_timeperiod_l'])
# LINEARREG_ANGLE - Linear Regression Angle
features['linearreg_angle_s'] = ta.LINEARREG_ANGLE(c, timeperiod=params['linearreg_angle_timeperiod_s'])
features['linearreg_angle_m'] = ta.LINEARREG_ANGLE(c, timeperiod=params['linearreg_angle_timeperiod_m'])
features['linearreg_angle_l'] = ta.LINEARREG_ANGLE(c, timeperiod=params['linearreg_angle_timeperiod_l'])
# LINEARREG_INTERCEPT - Linear Regression Intercept
features['linearreg_intercept_s'] = ta.LINEARREG_INTERCEPT(c, timeperiod=params['linearreg_intercept_timeperiod_s'])
features['linearreg_intercept_m'] = ta.LINEARREG_INTERCEPT(c, timeperiod=params['linearreg_intercept_timeperiod_m'])
features['linearreg_intercept_l'] = ta.LINEARREG_INTERCEPT(c, timeperiod=params['linearreg_intercept_timeperiod_l'])
# LINEARREG_SLOPE - Linear Regression Slope
features['linearreg_slope_s'] = ta.LINEARREG_SLOPE(c, timeperiod=params['linearreg_slope_timeperiod_s'])
features['linearreg_slope_m'] = ta.LINEARREG_SLOPE(c, timeperiod=params['linearreg_slope_timeperiod_m'])
features['linearreg_slope_l'] = ta.LINEARREG_SLOPE(c, timeperiod=params['linearreg_slope_timeperiod_l'])
# STDDEV - Standard Deviation
features['stddev_s'] = ta.STDDEV(c, timeperiod=params['stddev_timeperiod_s'], nbdev=1)
features['stddev_m'] = ta.STDDEV(c, timeperiod=params['stddev_timeperiod_m'], nbdev=1)
features['stddev_l'] = ta.STDDEV(c, timeperiod=params['stddev_timeperiod_l'], nbdev=1)
# TSF - Time Series Forecast
features['tsf_s'] = ta.TSF(c, timeperiod=params['tsf_timeperiod_s'])
features['tsf_m'] = ta.TSF(c, timeperiod=params['tsf_timeperiod_m'])
features['tsf_l'] = ta.TSF(c, timeperiod=params['tsf_timeperiod_l'])
# VAR - Variance
#features['var_s'] = ta.VAR(c, timeperiod=params['var_timeperiod_s'], nbdev=1)
#features['var_m'] = ta.VAR(c, timeperiod=params['var_timeperiod_m'], nbdev=1)
#features['var_l'] = ta.VAR(c, timeperiod=params['var_timeperiod_l'], nbdev=1)
# ボリンジャーバンド
# bbands_timeperiod_l = 30の場合、30分足で(30 * 30) / 60 = 15時間必要
bb_upper, bb_middle, bb_lower = ta.BBANDS(c, timeperiod=params['bbands_timeperiod_s'],
nbdevup=params['bbands_nbdevup_s'], nbdevdn=params['bbands_nbdevdn_s'],
matype=ta.MA_Type.EMA)
bb_trend1 = pd.Series(np.zeros(len(candle)), index=candle.index)
bb_trend1[c > bb_upper] = 1
bb_trend1[c < bb_lower] = -1
bb_trend2 = pd.Series(np.zeros(len(candle)), index=candle.index)
bb_trend2[c > bb_middle] = 1
bb_trend2[c < bb_middle] = -1
features['bb_upper_s'] = bb_upper
features['bb_middle_s'] = bb_middle
features['bb_lower_s'] = bb_lower
features['bb_trend1_s'] = bb_trend1
features['bb_trend2_s'] = bb_trend2
bb_upper, bb_middle, bb_lower = ta.BBANDS(c, timeperiod=params['bbands_timeperiod_m'],
nbdevup=params['bbands_nbdevup_m'], nbdevdn=params['bbands_nbdevdn_m'],
matype=ta.MA_Type.EMA)
bb_trend1 = pd.Series(np.zeros(len(candle)), index=candle.index)
bb_trend1[c > bb_upper] = 1
bb_trend1[c < bb_lower] = -1
bb_trend2 = pd.Series(np.zeros(len(candle)), index=candle.index)
bb_trend2[c > bb_middle] = 1
bb_trend2[c < bb_middle] = -1
features['bb_upper_m'] = bb_upper
features['bb_middle_m'] = bb_middle
features['bb_lower_m'] = bb_lower
features['bb_trend1_m'] = bb_trend1
features['bb_trend2_m'] = bb_trend2
bb_upper, bb_middle, bb_lower = ta.BBANDS(c, timeperiod=params['bbands_timeperiod_l'],
nbdevup=params['bbands_nbdevup_l'], nbdevdn=params['bbands_nbdevdn_l'],
matype=ta.MA_Type.EMA)
bb_trend1 = pd.Series(np.zeros(len(candle)), index=candle.index)
bb_trend1[c > bb_upper] = 1
bb_trend1[c < bb_lower] = -1
bb_trend2 = pd.Series(np.zeros(len(candle)), index=candle.index)
bb_trend2[c > bb_middle] = 1
bb_trend2[c < bb_middle] = -1
features['bb_upper_l'] = bb_upper
features['bb_middle_l'] = bb_middle
features['bb_lower_l'] = bb_lower
features['bb_trend1_l'] = bb_trend1
features['bb_trend2_l'] = bb_trend2
# ローソク足
features['CDL2CROWS'] = ta.CDL2CROWS(o, h, l, c)
features['CDL3BLACKCROWS'] = ta.CDL3BLACKCROWS(o, h, l, c)
features['CDL3INSIDE'] = ta.CDL3INSIDE(o, h, l, c)
features['CDL3LINESTRIKE'] = ta.CDL3LINESTRIKE(o, h, l, c)
features['CDL3OUTSIDE'] = ta.CDL3OUTSIDE(o, h, l, c)
features['CDL3STARSINSOUTH'] = ta.CDL3STARSINSOUTH(o, h, l, c)
features['CDL3WHITESOLDIERS'] = ta.CDL3WHITESOLDIERS(o, h, l, c)
features['CDLABANDONEDBABY'] = ta.CDLABANDONEDBABY(o, h, l, c, penetration=0)
features['CDLADVANCEBLOCK'] = ta.CDLADVANCEBLOCK(o, h, l, c)
features['CDLBELTHOLD'] = ta.CDLBELTHOLD(o, h, l, c)
features['CDLBREAKAWAY'] = ta.CDLBREAKAWAY(o, h, l, c)
features['CDLCLOSINGMARUBOZU'] = ta.CDLCLOSINGMARUBOZU(o, h, l, c)
features['CDLCONCEALBABYSWALL'] = ta.CDLCONCEALBABYSWALL(o, h, l, c)
features['CDLCOUNTERATTACK'] = ta.CDLCOUNTERATTACK(o, h, l, c)
features['CDLDARKCLOUDCOVER'] = ta.CDLDARKCLOUDCOVER(o, h, l, c, penetration=0)
#features['CDLDOJI'] = ta.CDLDOJI(o, h, l, c)
features['CDLDOJISTAR'] = ta.CDLDOJISTAR(o, h, l, c)
features['CDLDRAGONFLYDOJI'] = ta.CDLDRAGONFLYDOJI(o, h, l, c)
features['CDLENGULFING'] = ta.CDLENGULFING(o, h, l, c)
features['CDLEVENINGDOJISTAR'] = ta.CDLEVENINGDOJISTAR(o, h, l, c, penetration=0)
features['CDLEVENINGSTAR'] = ta.CDLEVENINGSTAR(o, h, l, c, penetration=0)
#features['CDLGAPSIDESIDEWHITE'] = ta.CDLGAPSIDESIDEWHITE(o, h, l, c)
features['CDLGRAVESTONEDOJI'] = ta.CDLGRAVESTONEDOJI(o, h, l, c)
features['CDLHAMMER'] = ta.CDLHAMMER(o, h, l, c)
#features['CDLHANGINGMAN'] = ta.CDLHANGINGMAN(o, h, l, c)
features['CDLHARAMI'] = ta.CDLHARAMI(o, h, l, c)
features['CDLHARAMICROSS'] = ta.CDLHARAMICROSS(o, h, l, c)
features['CDLHIGHWAVE'] = ta.CDLHIGHWAVE(o, h, l, c)
#features['CDLHIKKAKE'] = ta.CDLHIKKAKE(o, h, l, c)
features['CDLHIKKAKEMOD'] = ta.CDLHIKKAKEMOD(o, h, l, c)
features['CDLHOMINGPIGEON'] = ta.CDLHOMINGPIGEON(o, h, l, c)
#features['CDLIDENTICAL3CROWS'] = ta.CDLIDENTICAL3CROWS(o, h, l, c)
features['CDLINNECK'] = ta.CDLINNECK(o, h, l, c)
#features['CDLINVERTEDHAMMER'] = ta.CDLINVERTEDHAMMER(o, h, l, c)
features['CDLKICKING'] = ta.CDLKICKING(o, h, l, c)
features['CDLKICKINGBYLENGTH'] = ta.CDLKICKINGBYLENGTH(o, h, l, c)
features['CDLLADDERBOTTOM'] = ta.CDLLADDERBOTTOM(o, h, l, c)
#features['CDLLONGLEGGEDDOJI'] = ta.CDLLONGLEGGEDDOJI(o, h, l, c)
features['CDLMARUBOZU'] = ta.CDLMARUBOZU(o, h, l, c)
#features['CDLMATCHINGLOW'] = ta.CDLMATCHINGLOW(o, h, l, c)
features['CDLMATHOLD'] = ta.CDLMATHOLD(o, h, l, c, penetration=0)
features['CDLMORNINGDOJISTAR'] = ta.CDLMORNINGDOJISTAR(o, h, l, c, penetration=0)
features['CDLMORNINGSTAR'] = ta.CDLMORNINGSTAR(o, h, l, c, penetration=0)
features['CDLONNECK'] = ta.CDLONNECK(o, h, l, c)
features['CDLPIERCING'] = ta.CDLPIERCING(o, h, l, c)
features['CDLRICKSHAWMAN'] = ta.CDLRICKSHAWMAN(o, h, l, c)
features['CDLRISEFALL3METHODS'] = ta.CDLRISEFALL3METHODS(o, h, l, c)
features['CDLSEPARATINGLINES'] = ta.CDLSEPARATINGLINES(o, h, l, c)
#features['CDLSHOOTINGSTAR'] = ta.CDLSHOOTINGSTAR(o, h, l, c)
features['CDLSHORTLINE'] = ta.CDLSHORTLINE(o, h, l, c)
#features['CDLSPINNINGTOP'] = ta.CDLSPINNINGTOP(o, h, l, c)
features['CDLSTALLEDPATTERN'] = ta.CDLSTALLEDPATTERN(o, h, l, c)
features['CDLSTICKSANDWICH'] = ta.CDLSTICKSANDWICH(o, h, l, c)
features['CDLTAKURI'] = ta.CDLTAKURI(o, h, l, c)
features['CDLTASUKIGAP'] = ta.CDLTASUKIGAP(o, h, l, c)
features['CDLTHRUSTING'] = ta.CDLTHRUSTING(o, h, l, c)
features['CDLTRISTAR'] = ta.CDLTRISTAR(o, h, l, c)
features['CDLUNIQUE3RIVER'] = ta.CDLUNIQUE3RIVER(o, h, l, c)
features['CDLUPSIDEGAP2CROWS'] = ta.CDLUPSIDEGAP2CROWS(o, h, l, c)
features['CDLXSIDEGAP3METHODS'] = ta.CDLXSIDEGAP3METHODS(o, h, l, c)
'''
# トレンドライン
for dt in datetimerange(candle.index[0], candle.index[-1] + timedelta(minutes=1)):
start_dt = (dt - timedelta(minutes=130)).strftime('%Y-%m-%d %H:%M:%S')
end_dt = dt.strftime('%Y-%m-%d %H:%M:%S')
tmp = candle.loc[(start_dt <= candle.index) & (candle.index <= end_dt)]
for w_size, stride in [(15, 5), (30, 10), (60, 10), (120, 10)]:
# for w_size, stride in [(120, 10)]:
trendlines = calc_trendlines(tmp, w_size, stride)
if len(trendlines) == 0:
continue
trendline_feature = calc_trendline_feature(tmp, trendlines)
print('{}-{} {} {} {}'.format(dt - timedelta(minutes=130), dt, trendline_feature['high_a'], trendline_feature['high_b'], trendline_feature['high_diff']))
features.loc[features.index == end_dt, 'trendline_high_a_{}'.format(w_size)] = trendline_feature['high_a']
features.loc[features.index == end_dt, 'trendline_high_b_{}'.format(w_size)] = trendline_feature['high_b']
features.loc[features.index == end_dt, 'trendline_high_diff_{}'.format(w_size)] = trendline_feature['high_diff']
features.loc[features.index == end_dt, 'trendline_low_a_{}'.format(w_size)] = trendline_feature['low_a']
features.loc[features.index == end_dt, 'trendline_low_b_{}'.format(w_size)] = trendline_feature['low_b']
features.loc[features.index == end_dt, 'trendline_low_diff_{}'.format(w_size)] = trendline_feature['low_diff']
'''
window = 5
features_ext = features
for w in range(window):
tmp = features.shift(periods=60 * (w + 1), freq='S')
tmp.columns = [c + '_' + str(w + 1) + 'w' for c in features.columns]
features_ext = pd.concat([features_ext, tmp], axis=1)
if candle_type == '5min':
features_ext = features_ext.shift(periods=300, freq='S')
features_ext = features_ext.fillna(method='ffill')
features_ext = features_ext[features_ext.index == target_dt]
return features_ext
def ta(name, price_h, price_l, price_c, price_v, price_o):
# function 'MAX'/'MAXINDEX'/'MIN'/'MININDEX'/'MINMAX'/'MINMAXINDEX'/'SUM' is missing
if name == 'AD':
return talib.AD(np.array(price_h), np.array(price_l),
np.array(price_c), np.asarray(price_v, dtype='float'))
if name == 'ADOSC':
return talib.ADOSC(np.array(price_h),
np.array(price_l),
np.array(price_c),
np.asarray(price_v, dtype='float'),
fastperiod=2,
slowperiod=10)
if name == 'ADX':
return talib.ADX(np.array(price_h),
np.array(price_l),
np.asarray(price_c, dtype='float'),
timeperiod=14)
if name == 'ADXR':
return talib.ADXR(np.array(price_h),
np.array(price_l),
np.asarray(price_c, dtype='float'),
timeperiod=14)
if name == 'APO':
return talib.APO(np.asarray(price_c, dtype='float'),
fastperiod=12,
slowperiod=26,
matype=0)
if name == 'AROON':
AROON_DWON, AROON2_UP = talib.AROON(np.array(price_h),
np.asarray(price_l, dtype='float'),
timeperiod=90)
return (AROON_DWON, AROON2_UP)
if name == 'AROONOSC':
return talib.AROONOSC(np.array(price_h),
np.asarray(price_l, dtype='float'),
timeperiod=14)
if name == 'ATR':
return talib.ATR(np.array(price_h),
np.array(price_l),
np.asarray(price_c, dtype='float'),
timeperiod=14)
if name == 'AVGPRICE':
return talib.AVGPRICE(np.array(price_o), np.array(price_h),
np.array(price_l),
np.asarray(price_c, dtype='float'))
if name == 'BBANDS':
BBANDS1, BBANDS2, BBANDS3 = talib.BBANDS(np.asarray(price_c,
dtype='float'),
matype=MA_Type.T3)
return BBANDS1
if name == 'BETA':
return talib.BETA(np.array(price_h),
np.asarray(price_l, dtype='float'),
timeperiod=5)
if name == 'BOP':
return talib.BOP(np.array(price_o), np.array(price_h),
np.array(price_l), np.asarray(price_c, dtype='float'))
if name == 'CCI':
return talib.CCI(np.array(price_h),
np.array(price_l),
np.asarray(price_c, dtype='float'),
timeperiod=14)
if name == 'CDL2CROWS':
return talib.CDL2CROWS(np.array(price_o), np.array(price_h),
np.array(price_l),
np.asarray(price_c, dtype='float'))
if name == 'CDL3BLACKCROWS':
return talib.CDL3BLACKCROWS(np.array(price_o), np.array(price_h),
np.array(price_l),
np.asarray(price_c, dtype='float'))
if name == 'CDL3INSIDE':
return talib.CDL3INSIDE(np.array(price_o), np.array(price_h),
np.array(price_l),
np.asarray(price_c, dtype='float'))
if name == 'CDL3LINESTRIKE':
return talib.CDL3LINESTRIKE(np.array(price_o), np.array(price_h),
np.array(price_l),
np.asarray(price_c, dtype='float'))
if name == 'CDL3OUTSIDE':
return talib.CDL3OUTSIDE(np.array(price_o), np.array(price_h),
np.array(price_l),
np.asarray(price_c, dtype='float'))
if name == 'CDL3STARSINSOUTH':
return talib.CDL3STARSINSOUTH(np.array(price_o), np.array(price_h),
np.array(price_l),
np.asarray(price_c, dtype='float'))
if name == 'CDL3WHITESOLDIERS':
return talib.CDL3WHITESOLDIERS(np.array(price_o), np.array(price_h),
np.array(price_l),
np.asarray(price_c, dtype='float'))
if name == 'CDLABANDONEDBABY':
return talib.CDLABANDONEDBABY(np.array(price_o),
np.array(price_h),
np.array(price_l),
np.asarray(price_c, dtype='float'),
penetration=0)
if name == 'CDLADVANCEBLOCK':
return talib.CDLADVANCEBLOCK(np.array(price_o), np.array(price_h),
np.array(price_l),
np.asarray(price_c, dtype='float'))
if name == 'CDLBELTHOLD':
return talib.CDLBELTHOLD(np.array(price_o), np.array(price_h),
np.array(price_l),
np.asarray(price_c, dtype='float'))
if name == 'CDLBREAKAWAY':
return talib.CDLBREAKAWAY(np.array(price_o), np.array(price_h),
np.array(price_l),
np.asarray(price_c, dtype='float'))
if name == 'CDLCLOSINGMARUBOZU':
return talib.CDLCLOSINGMARUBOZU(np.array(price_o), np.array(price_h),
np.array(price_l),
np.asarray(price_c, dtype='float'))
if name == 'CDLCONCEALBABYSWALL':
return talib.CDLCONCEALBABYSWALL(np.array(price_o), np.array(price_h),
np.array(price_l),
np.asarray(price_c, dtype='float'))
if name == 'CDLCOUNTERATTACK':
return talib.CDLCOUNTERATTACK(np.array(price_o), np.array(price_h),
np.array(price_l),
np.asarray(price_c, dtype='float'))
if name == 'CDLDARKCLOUDCOVER':
return talib.CDLDARKCLOUDCOVER(np.array(price_o),
np.array(price_h),
np.array(price_l),
np.asarray(price_c, dtype='float'),
penetration=0)
if name == 'CDLDOJI':
return talib.CDLDOJI(np.array(price_o), np.array(price_h),
np.array(price_l),
np.asarray(price_c, dtype='float'))
if name == 'CDLDOJISTAR':
return talib.CDLDOJISTAR(np.array(price_o), np.array(price_h),
np.array(price_l),
np.asarray(price_c, dtype='float'))
if name == 'CDLDRAGONFLYDOJI':
return talib.CDLDRAGONFLYDOJI(np.array(price_o), np.array(price_h),
np.array(price_l),
np.asarray(price_c, dtype='float'))
if name == 'CDLENGULFING':
return talib.CDLENGULFING(np.array(price_o), np.array(price_h),
np.array(price_l),
np.asarray(price_c, dtype='float'))
if name == 'CDLEVENINGDOJISTAR':
return talib.CDLEVENINGDOJISTAR(np.array(price_o),
np.array(price_h),
np.array(price_l),
np.asarray(price_c, dtype='float'),
penetration=0)
if name == 'CDLEVENINGSTAR':
return talib.CDLEVENINGSTAR(np.array(price_o),
np.array(price_h),
np.array(price_l),
np.asarray(price_c, dtype='float'),
penetration=0)
if name == 'CDLGAPSIDESIDEWHITE':
return talib.CDLGAPSIDESIDEWHITE(np.array(price_o), np.array(price_h),
np.array(price_l),
np.asarray(price_c, dtype='float'))
if name == 'CDLGRAVESTONEDOJI':
return talib.CDLGRAVESTONEDOJI(np.array(price_o), np.array(price_h),
np.array(price_l),
np.asarray(price_c, dtype='float'))
if name == 'CDLHAMMER':
return talib.CDLHAMMER(np.array(price_o), np.array(price_h),
np.array(price_l),
np.asarray(price_c, dtype='float'))
if name == 'CDLHANGINGMAN':
return talib.CDLHANGINGMAN(np.array(price_o), np.array(price_h),
np.array(price_l),
np.asarray(price_c, dtype='float'))
if name == 'CDLHARAMI':
return talib.CDLHARAMI(np.array(price_o), np.array(price_h),
np.array(price_l),
np.asarray(price_c, dtype='float'))
if name == 'CDLHARAMICROSS':
return talib.CDLHARAMICROSS(np.array(price_o), np.array(price_h),
np.array(price_l),
np.asarray(price_c, dtype='float'))
if name == 'CDLHIGHWAVE':
return talib.CDLHIGHWAVE(np.array(price_o), np.array(price_h),
np.array(price_l),
np.asarray(price_c, dtype='float'))
if name == 'CDLHIKKAKE':
return talib.CDLHIKKAKE(np.array(price_o), np.array(price_h),
np.array(price_l),
np.asarray(price_c, dtype='float'))
if name == 'CDLHIKKAKEMOD':
return talib.CDLHIKKAKEMOD(np.array(price_o), np.array(price_h),
np.array(price_l),
np.asarray(price_c, dtype='float'))
if name == 'CDLHOMINGPIGEON':
return talib.CDLHOMINGPIGEON(np.array(price_o), np.array(price_h),
np.array(price_l),
np.asarray(price_c, dtype='float'))
if name == 'CDLIDENTICAL3CROWS':
return talib.CDLIDENTICAL3CROWS(np.array(price_o), np.array(price_h),
np.array(price_l),
np.asarray(price_c, dtype='float'))
if name == 'CDLINNECK':
return talib.CDLINNECK(np.array(price_o), np.array(price_h),
np.array(price_l),
np.asarray(price_c, dtype='float'))
if name == 'CDLINVERTEDHAMMER':
return talib.CDLINVERTEDHAMMER(np.array(price_o), np.array(price_h),
np.array(price_l),
np.asarray(price_c, dtype='float'))
if name == 'CDLKICKING':
return talib.CDLKICKING(np.array(price_o), np.array(price_h),
np.array(price_l),
np.asarray(price_c, dtype='float'))
if name == 'CDLKICKINGBYLENGTH':
return talib.CDLKICKINGBYLENGTH(np.array(price_o), np.array(price_h),
np.array(price_l),
np.asarray(price_c, dtype='float'))
if name == 'CDLLADDERBOTTOM':
return talib.CDLLADDERBOTTOM(np.array(price_o), np.array(price_h),
np.array(price_l),
np.asarray(price_c, dtype='float'))
if name == 'CDLLONGLEGGEDDOJI':
return talib.CDLLONGLEGGEDDOJI(np.array(price_o), np.array(price_h),
np.array(price_l),
np.asarray(price_c, dtype='float'))
if name == 'CDLLONGLINE':
return talib.CDLLONGLINE(np.array(price_o), np.array(price_h),
np.array(price_l),
np.asarray(price_c, dtype='float'))
if name == 'CDLMARUBOZU':
return talib.CDLMARUBOZU(np.array(price_o), np.array(price_h),
np.array(price_l),
np.asarray(price_c, dtype='float'))
if name == 'CDLMATCHINGLOW':
return talib.CDLMATCHINGLOW(np.array(price_o), np.array(price_h),
np.array(price_l),
np.asarray(price_c, dtype='float'))
if name == 'CDLMATHOLD':
return talib.CDLMATHOLD(np.array(price_o), np.array(price_h),
np.array(price_l),
np.asarray(price_c, dtype='float'))
if name == 'CDLMORNINGDOJISTAR':
return talib.CDLMORNINGDOJISTAR(np.array(price_o),
np.array(price_h),
np.array(price_l),
np.asarray(price_c, dtype='float'),
penetration=0)
if name == 'CDLMORNINGSTAR':
return talib.CDLMORNINGSTAR(np.array(price_o),
np.array(price_h),
np.array(price_l),
np.asarray(price_c, dtype='float'),
penetration=0)
if name == 'CDLONNECK':
return talib.CDLONNECK(np.array(price_o), np.array(price_h),
np.array(price_l),
np.asarray(price_c, dtype='float'))
if name == 'CDLPIERCING':
return talib.CDLPIERCING(np.array(price_o), np.array(price_h),
np.array(price_l),
np.asarray(price_c, dtype='float'))
if name == 'CDLRICKSHAWMAN':
return talib.CDLRICKSHAWMAN(np.array(price_o), np.array(price_h),
np.array(price_l),
np.asarray(price_c, dtype='float'))
if name == 'CDLRISEFALL3METHODS':
return talib.CDLRISEFALL3METHODS(np.array(price_o), np.array(price_h),
np.array(price_l),
np.asarray(price_c, dtype='float'))
if name == 'CDLSEPARATINGLINES':
return talib.CDLSEPARATINGLINES(np.array(price_o), np.array(price_h),
np.array(price_l),
np.asarray(price_c, dtype='float'))
if name == 'CDLSHOOTINGSTAR':
return talib.CDLSHOOTINGSTAR(np.array(price_o), np.array(price_h),
np.array(price_l),
np.asarray(price_c, dtype='float'))
if name == 'CDLSHORTLINE':
return talib.CDLSHORTLINE(np.array(price_o), np.array(price_h),
np.array(price_l),
np.asarray(price_c, dtype='float'))
if name == 'CDLSPINNINGTOP':
return talib.CDLSPINNINGTOP(np.array(price_o), np.array(price_h),
np.array(price_l),
np.asarray(price_c, dtype='float'))
if name == 'CDLSTALLEDPATTERN':
return talib.CDLSTALLEDPATTERN(np.array(price_o), np.array(price_h),
np.array(price_l),
np.asarray(price_c, dtype='float'))
if name == 'CDLSTICKSANDWICH':
return talib.CDLSTICKSANDWICH(np.array(price_o), np.array(price_h),
np.array(price_l),
np.asarray(price_c, dtype='float'))
if name == 'CDLTAKURI':
return talib.CDLTAKURI(np.array(price_o), np.array(price_h),
np.array(price_l),
np.asarray(price_c, dtype='float'))
if name == 'CDLTASUKIGAP':
return talib.CDLTASUKIGAP(np.array(price_o), np.array(price_h),
np.array(price_l),
np.asarray(price_c, dtype='float'))
if name == 'CDLTHRUSTING':
return talib.CDLTHRUSTING(np.array(price_o), np.array(price_h),
np.array(price_l),
np.asarray(price_c, dtype='float'))
if name == 'CDLTRISTAR':
return talib.CDLTRISTAR(np.array(price_o), np.array(price_h),
np.array(price_l),
np.asarray(price_c, dtype='float'))
if name == 'CDLUNIQUE3RIVER':
return talib.CDLUNIQUE3RIVER(np.array(price_o), np.array(price_h),
np.array(price_l),
np.asarray(price_c, dtype='float'))
if name == 'CDLUPSIDEGAP2CROWS':
return talib.CDLUPSIDEGAP2CROWS(np.array(price_o), np.array(price_h),
np.array(price_l),
np.asarray(price_c, dtype='float'))
if name == 'CDLXSIDEGAP3METHODS':
return talib.CDLXSIDEGAP3METHODS(np.array(price_o), np.array(price_h),
np.array(price_l),
np.asarray(price_c, dtype='float'))
if name == 'CMO':
return talib.CMO(np.asarray(price_c, dtype='float'), timeperiod=14)
if name == 'CORREL':
return talib.CORREL(np.array(price_h),
np.asarray(price_l, dtype='float'),
timeperiod=30)
if name == 'DEMA':
return talib.DEMA(np.asarray(price_c, dtype='float'), timeperiod=30)
if name == 'DX':
return talib.DX(np.array(price_h),
np.array(price_l),
np.asarray(price_c, dtype='float'),
timeperiod=14)
if name == 'EMA':
return talib.EMA(np.asarray(price_c, dtype='float'), timeperiod=30)
if name == 'HT_DCPERIOD':
return talib.HT_DCPERIOD(np.asarray(price_c, dtype='float'))
if name == 'HT_DCPHASE':
return talib.HT_DCPHASE(np.asarray(price_c, dtype='float'))
if name == 'HT_PHASOR':
HT_PHASOR1, HT_PHASOR2 = talib.HT_PHASOR(
np.asarray(price_c, dtype='float')
) # use HT_PHASOR1 for the indication of up and down
return HT_PHASOR1
if name == 'HT_SINE':
HT_SINE1, HT_SINE2 = talib.HT_SINE(np.asarray(price_c, dtype='float'))
return HT_SINE1
if name == 'HT_TRENDLINE':
return talib.HT_TRENDLINE(np.asarray(price_c, dtype='float'))
if name == 'HT_TRENDMODE':
return talib.HT_TRENDMODE(np.asarray(price_c, dtype='float'))
if name == 'KAMA':
return talib.KAMA(np.asarray(price_c, dtype='float'), timeperiod=30)
if name == 'LINEARREG':
return talib.LINEARREG(np.asarray(price_c, dtype='float'),
timeperiod=14)
if name == 'LINEARREG_ANGLE':
return talib.LINEARREG_ANGLE(np.asarray(price_c, dtype='float'),
timeperiod=14)
if name == 'LINEARREG_INTERCEPT':
return talib.LINEARREG_INTERCEPT(np.asarray(price_c, dtype='float'),
timeperiod=14)
if name == 'LINEARREG_SLOPE':
return talib.LINEARREG_SLOPE(np.asarray(price_c, dtype='float'),
timeperiod=14)
if name == 'MA':
return talib.MA(np.asarray(price_c, dtype='float'),
timeperiod=30,
matype=0)
if name == 'MACD':
MACD1, MACD2, MACD3 = talib.MACD(np.asarray(price_c, dtype='float'),
fastperiod=12,
slowperiod=26,
signalperiod=9)
return MACD1
if nam == 'MACDEXT':
return talib.MACDEXT(np.asarray(price_c, dtype='float'),
fastperiod=12,
fastmatype=0,
slowperiod=26,
slowmatype=0,
signalperiod=9,
signalmatype=0)
if name == 'MACDFIX':
MACDFIX1, MACDFIX2, MACDFIX3 = talib.MACDFIX(np.asarray(price_c,
dtype='float'),
signalperiod=9)
return MACDFIX1
if name == 'MAMA':
MAMA1, MAMA2 = talib.MAMA(np.asarray(price_c, dtype='float'),
fastlimit=0,
slowlimit=0)
return MAMA1
if name == 'MEDPRICE':
return talib.MEDPRICE(np.array(price_h),
np.asarray(price_l, dtype='float'))
if name == 'MINUS_DI':
return talib.MINUS_DI(np.array(price_h),
np.array(price_l),
np.asarray(price_c, dtype='float'),
timeperiod=14)
if name == 'MINUS_DM':
return talib.MINUS_DM(np.array(price_h),
np.asarray(price_l, dtype='float'),
timeperiod=14)
if name == 'MOM':
return talib.MOM(np.asarray(price_c, dtype='float'), timeperiod=10)
if name == 'NATR':
return talib.NATR(np.array(price_h),
np.array(price_l),
np.asarray(price_c, dtype='float'),
timeperiod=14)
if name == 'OBV':
return talib.OBV(np.array(price_c), np.asarray(price_v, dtype='float'))
if name == 'PLUS_DI':
return talib.PLUS_DI(np.array(price_h),
np.array(price_l),
np.asarray(price_c, dtype='float'),
timeperiod=14)
if name == 'PLUS_DM':
return talib.PLUS_DM(np.array(price_h),
np.asarray(price_l, dtype='float'),
timeperiod=14)
if name == 'PPO':
return talib.PPO(np.asarray(price_c, dtype='float'),
fastperiod=12,
slowperiod=26,
matype=0)
if name == 'ROC':
return talib.ROC(np.asarray(price_c, dtype='float'), timeperiod=10)
if name == 'ROCP':
return talib.ROCP(np.asarray(price_c, dtype='float'), timeperiod=10)
if name == 'ROCR100':
return talib.ROCR100(np.asarray(price_c, dtype='float'), timeperiod=10)
if name == 'RSI':
return talib.RSI(np.asarray(price_c, dtype='float'), timeperiod=14)
if name == 'SAR':
return talib.SAR(np.array(price_h),
np.asarray(price_l, dtype='float'),
acceleration=0,
maximum=0)
if name == 'SAREXT':
return talib.SAREXT(np.array(price_h),
np.asarray(price_l, dtype='float'),
startvalue=0,
offsetonreverse=0,
accelerationinitlong=0,
accelerationlong=0,
accelerationmaxlong=0,
accelerationinitshort=0,
accelerationshort=0,
accelerationmaxshort=0)
if name == 'SMA':
return talib.SMA(np.asarray(price_c, dtype='float'), timeperiod=30)
if name == 'STDDEV':
return talib.STDDEV(np.asarray(price_c, dtype='float'),
timeperiod=5,
nbdev=1)
if name == 'STOCH':
STOCH1, STOCH2 = talib.STOCH(np.array(price_h),
np.array(price_l),
np.asarray(price_c, dtype='float'),
fastk_period=5,
slowk_period=3,
slowk_matype=0,
slowd_period=3,
slowd_matype=0)
return STOCH1
if name == 'STOCHF':
STOCHF1, STOCHF2 = talib.STOCHF(np.array(price_h),
np.array(price_l),
np.asarray(price_c, dtype='float'),
fastk_period=5,
fastd_period=3,
fastd_matype=0)
return STOCHF1
if name == 'STOCHRSI':
STOCHRSI1, STOCHRSI2 = talib.STOCHRSI(np.asarray(price_c,
dtype='float'),
timeperiod=14,
fastk_period=5,
fastd_period=3,
fastd_matype=0)
return STOCHRSI1
if name == 'T3':
return talib.T3(np.asarray(price_c, dtype='float'),
timeperiod=5,
vfactor=0)
if name == 'TEMA':
return talib.TEMA(np.asarray(price_c, dtype='float'), timeperiod=30)
if name == 'TRANGE':
return talib.TRANGE(np.array(price_h), np.array(price_l),
np.asarray(price_c, dtype='float'))
if name == 'TRIMA':
return talib.TRIMA(np.asarray(price_c, dtype='float'), timeperiod=30)
if name == 'TRIX':
return talib.TRIX(np.asarray(price_c, dtype='float'), timeperiod=30)
if name == 'TSF':
return talib.TSF(np.asarray(price_c, dtype='float'), timeperiod=14)
if name == 'TYPPRICE':
return talib.TYPPRICE(np.array(price_h), np.array(price_l),
np.asarray(price_c, dtype='float'))
if name == 'ULTOSC':
return talib.ULTOSC(np.array(price_h),
np.array(price_l),
np.asarray(price_c, dtype='float'),
timeperiod1=7,
timeperiod2=14,
timeperiod3=28)
if name == 'VAR':
return talib.VAR(np.asarray(price_c, dtype='float'),
timeperiod=5,
nbdev=1)
if name == 'WCLPRICE':
return talib.WCLPRICE(np.array(price_h), np.array(price_l),
np.asarray(price_c, dtype='float'))
if name == 'WILLR':
return talib.WILLR(np.array(price_h),
np.array(price_l),
np.asarray(price_c, dtype='float'),
timeperiod=14)
if name == 'WMA':
return talib.WMA(np.asarray(price_c, dtype='float'), timeperiod=30)
async def indicators(exchange: str,
symbol: str,
interval: str = '30m',
limit: int = 100):
exchange = getattr(ccxt, exchange)()
kline = exchange.fetch_ohlcv(symbol, interval, limit=int(limit))
df = pd.DataFrame(
np.array(kline),
columns=['open_time', 'open', 'high', 'low', 'close', 'volume'],
dtype='float64')
op = df['open']
hi = df['high']
lo = df['low']
cl = df['close']
vl = df['volume']
adx = talib.ADX(hi.values, lo.values, cl.values, timeperiod=14)
rsi = talib.RSI(cl.values, timeperiod=14)
plus_di = talib.PLUS_DI(hi.values, lo.values, cl.values, timeperiod=14)
minus_di = talib.MINUS_DI(hi.values, lo.values, cl.values, timeperiod=14)
sma = talib.SMA(cl.values, timeperiod=30)
sma_5 = talib.SMA(cl.values, timeperiod=5)
sma_10 = talib.SMA(cl.values, timeperiod=10)
sma_dir = convert_number(round(sma[-1], 8) - round(sma_10[-1], 8))
macd, macdsignal, macdhist = talib.MACD(cl.values,
fastperiod=12,
slowperiod=26,
signalperiod=14)
macd = convert_number(macd[-1])
macdsignal = convert_number(macdsignal[-1])
ma_50 = convert_number(talib.MA(cl.values, timeperiod=50, matype=0)[-1])
ma_100 = convert_number(talib.MA(cl.values, timeperiod=100, matype=0)[-1])
obv = talib.OBV(cl.values, vl.values)
rsi_obv = convert_number(talib.RSI(obv, timeperiod=14)[-1])
linear_regression = talib.LINEARREG(cl.values, timeperiod=14)[-1]
linear_angle = convert_number(
talib.LINEARREG_ANGLE(cl.values, timeperiod=14)[-1])
linear_intercept = convert_number(
talib.LINEARREG_INTERCEPT(cl.values, timeperiod=14)[-1])
linear_slope = convert_number(
talib.LINEARREG_SLOPE(cl.values, timeperiod=14)[-1])
return {
"adx": adx[-1],
"rsi": rsi[-1],
"plus_di": plus_di[-1],
"minus_di": minus_di[-1],
"sma": round(sma[-1], 8),
"sma_10": round(sma_10[-1], 8),
"sma_5": round(sma_5[-1], 8),
"sma_dir": sma_dir,
"macd": macd,
"macdsignal": macdsignal,
"ma_50": ma_50,
"ma_100": ma_100,
"rsi_obv": rsi_obv,
"linear_regression": linear_regression,
"linear_angle": linear_angle,
"linear_intercept": linear_intercept,
"linear_slope": linear_slope
}
def LINEARREG_ANGLE(Series, N=14):
res = talib.LINEARREG_ANGLE(Series.values, N)
return pd.Series(res, index=Series.index)
def calc_features(df):
open = df['op']
high = df['hi']
low = df['lo']
close = df['cl']
volume = df['volume']
orig_columns = df.columns
hilo = (df['hi'] + df['lo']) / 2
df['BBANDS_upperband'], df['BBANDS_middleband'], df[
'BBANDS_lowerband'] = talib.BBANDS(close,
timeperiod=5,
nbdevup=2,
nbdevdn=2,
matype=0)
df['BBANDS_upperband'] -= hilo
df['BBANDS_middleband'] -= hilo
df['BBANDS_lowerband'] -= hilo
df['DEMA'] = talib.DEMA(close, timeperiod=30) - hilo
df['EMA'] = talib.EMA(close, timeperiod=30) - hilo
df['HT_TRENDLINE'] = talib.HT_TRENDLINE(close) - hilo
df['KAMA'] = talib.KAMA(close, timeperiod=30) - hilo
df['MA'] = talib.MA(close, timeperiod=30, matype=0) - hilo
df['MIDPOINT'] = talib.MIDPOINT(close, timeperiod=14) - hilo
df['SMA'] = talib.SMA(close, timeperiod=30) - hilo
df['T3'] = talib.T3(close, timeperiod=5, vfactor=0) - hilo
df['TEMA'] = talib.TEMA(close, timeperiod=30) - hilo
df['TRIMA'] = talib.TRIMA(close, timeperiod=30) - hilo
df['WMA'] = talib.WMA(close, timeperiod=30) - hilo
df['ADX'] = talib.ADX(high, low, close, timeperiod=14)
df['ADXR'] = talib.ADXR(high, low, close, timeperiod=14)
df['APO'] = talib.APO(close, fastperiod=12, slowperiod=26, matype=0)
df['AROON_aroondown'], df['AROON_aroonup'] = talib.AROON(high,
low,
timeperiod=14)
df['AROONOSC'] = talib.AROONOSC(high, low, timeperiod=14)
df['BOP'] = talib.BOP(open, high, low, close)
df['CCI'] = talib.CCI(high, low, close, timeperiod=14)
df['DX'] = talib.DX(high, low, close, timeperiod=14)
df['MACD_macd'], df['MACD_macdsignal'], df['MACD_macdhist'] = talib.MACD(
close, fastperiod=12, slowperiod=26, signalperiod=9)
# skip MACDEXT MACDFIX たぶん同じなので
df['MFI'] = talib.MFI(high, low, close, volume, timeperiod=14)
df['MINUS_DI'] = talib.MINUS_DI(high, low, close, timeperiod=14)
df['MINUS_DM'] = talib.MINUS_DM(high, low, timeperiod=14)
df['MOM'] = talib.MOM(close, timeperiod=10)
df['PLUS_DI'] = talib.PLUS_DI(high, low, close, timeperiod=14)
df['PLUS_DM'] = talib.PLUS_DM(high, low, timeperiod=14)
df['RSI'] = talib.RSI(close, timeperiod=14)
df['STOCH_slowk'], df['STOCH_slowd'] = talib.STOCH(high,
low,
close,
fastk_period=5,
slowk_period=3,
slowk_matype=0,
slowd_period=3,
slowd_matype=0)
df['STOCHF_fastk'], df['STOCHF_fastd'] = talib.STOCHF(high,
low,
close,
fastk_period=5,
fastd_period=3,
fastd_matype=0)
df['STOCHRSI_fastk'], df['STOCHRSI_fastd'] = talib.STOCHRSI(close,
timeperiod=14,
fastk_period=5,
fastd_period=3,
fastd_matype=0)
df['TRIX'] = talib.TRIX(close, timeperiod=30)
df['ULTOSC'] = talib.ULTOSC(high,
low,
close,
timeperiod1=7,
timeperiod2=14,
timeperiod3=28)
df['WILLR'] = talib.WILLR(high, low, close, timeperiod=14)
df['AD'] = talib.AD(high, low, close, volume)
df['ADOSC'] = talib.ADOSC(high,
low,
close,
volume,
fastperiod=3,
slowperiod=10)
df['OBV'] = talib.OBV(close, volume)
df['ATR'] = talib.ATR(high, low, close, timeperiod=14)
df['NATR'] = talib.NATR(high, low, close, timeperiod=14)
df['TRANGE'] = talib.TRANGE(high, low, close)
df['HT_DCPERIOD'] = talib.HT_DCPERIOD(close)
df['HT_DCPHASE'] = talib.HT_DCPHASE(close)
df['HT_PHASOR_inphase'], df['HT_PHASOR_quadrature'] = talib.HT_PHASOR(
close)
df['HT_SINE_sine'], df['HT_SINE_leadsine'] = talib.HT_SINE(close)
df['HT_TRENDMODE'] = talib.HT_TRENDMODE(close)
df['BETA'] = talib.BETA(high, low, timeperiod=5)
df['CORREL'] = talib.CORREL(high, low, timeperiod=30)
df['LINEARREG'] = talib.LINEARREG(close, timeperiod=14) - close
df['LINEARREG_ANGLE'] = talib.LINEARREG_ANGLE(close, timeperiod=14)
df['LINEARREG_INTERCEPT'] = talib.LINEARREG_INTERCEPT(
close, timeperiod=14) - close
df['LINEARREG_SLOPE'] = talib.LINEARREG_SLOPE(close, timeperiod=14)
df['STDDEV'] = talib.STDDEV(close, timeperiod=5, nbdev=1)
return df
turnoverx = np.asarray(datax['turnover'])
roc1 = ta.ROC(closex, 1)
roc5 = ta.ROC(closex, 5)
roc10 = ta.ROC(closex, 10)
rsi = ta.RSI(closex, 14)
ultosc = ta.ULTOSC(highx, lowx, closex)
slowk, slowd = ta.STOCH(highx, lowx, closex)
slow_diff = slowk - slowd
adx = ta.ADX(highx, lowx, closex, 14)
ppo = ta.PPO(closex)
willr = ta.WILLR(highx, lowx, closex)
cci = ta.CCI(highx, lowx, closex)
bop = ta.BOP(openx, highx, lowx, closex)
std_p20 = ta.STDDEV(closex, 20)
angle = ta.LINEARREG_ANGLE(closex)
mfi = ta.MFI(highx, lowx, closex, volumex)
adosc = ta.ADOSC(highx, lowx, closex, turnoverx)
## obtian training data
start_day = '20150105'
end_day = '20150714'
st = (datax['date'] == start_day).nonzero()[0][0]
ed = (datax['date'] == end_day).nonzero()[0][0]
features = np.column_stack((roc1, roc5, roc10,rsi,ultosc,slowk,slow_diff,adx,ppo,willr,\
cci,bop,std_p20,angle,mfi,adosc))
features2 = np.float64(features[st:ed, :])
