def loadIndex(self):
data=pd.read_csv(self.indexPath,engine='python')
data=data[-self.date_duration-1:-1]
data=data.reset_index()
data=data.fillna(method='ffill')
data=np.array(data['Open'])
data=np.reshape(data,(-1,1))
data=self.indexScaler.fit_transform(data)
data=pd.DataFrame(np.squeeze(data))
print('IndexPrice shape: ',data.shape)
return data
def BiznesRadarParse(content):
''' Parse file HTML content to get stocks table '''
table = GetHTMLElement(content, 'table', 'qTableFull')
#DebugSave('table.html', table)
data = pd.read_html(table,
thousands=' ',
decimal='.',
displayed_only=False)[0]
# Remove separator lines
data = data[data.ROE != 'ROE']
data = data.reset_index()
data = data.rename(
columns={
'Cena / Wartość księgowa': 'C/WK',
'Cena / Przychody ze sprzedaży': 'C/P',
'Cena / Zysk': 'C/Z',
'Aktualny kurs': 'Kurs',
'Średni obrót z 5 sesji [zł]': 'Obrot',
'Piotroski F-Score': 'Piotroski',
'Trend 6m': 'T6M',
'Trend 12m': 'T12M',
'Trend 24m': 'T24M',
'Zmiana kursu 3m [%]': 'Z3',
'Zmiana kursu 6m [%]': 'Z6',
'Zmiana kursu 12m [%]': 'Z12'
})
# Convert string values to float/int values
for i in (range(len(data['Profil']))):
data['ROE'][i] = float(data['ROE'][i].replace(' ', '').strip('%'))
data['ROA'][i] = float(data['ROA'][i].replace(' ', '').strip('%'))
data['C/WK'][i] = float(data['C/WK'][i])
data['C/P'][i] = float(data['C/P'][i])
data['C/Z'][i] = float(data['C/Z'][i])
data['Kurs'][i] = float(data['Kurs'][i])
data['Obrot'][i] = int(data['Obrot'][i])
data['Piotroski'][i] = int(data['Piotroski'][i])
data['Z3'][i] = float(data['Z3'][i].replace(' ', '').strip('%'))
data['Z6'][i] = float(data['Z6'][i].replace(' ', '').strip('%'))
data['Z12'][i] = float(data['Z12'][i].replace(' ', '').strip('%'))
return data
def prophet(data):
"""takes in df and produces df of results"""
from fbprophet import Prophet
data = data.reset_index()
data = data.rename(columns={'index': 'ds', 'close': 'y'})
prophet = Prophet(interval_width=0.95)
prophet.fit(data)
future = prophet.make_future_dataframe(periods=(365), freq='D')
forecast = prophet.predict(future)
results_df = forecast[['ds', 'yhat', 'yhat_lower',
'yhat_upper']].iloc[-365:]
results_df.rename(columns={
'ds': 'date',
'yhat': 'predicted',
'yhat_lower': 'lower_bound',
'yhat_upper': 'upper_bound'
},
inplace=True)
results_df.set_index('date', inplace=True)
return results_df
def loadCSV(self):
"""
csv파일이 있는 폴더를 입력으로 받아 데이터를 읽어옴.
"""
csvList=os.listdir(self.folderPath)
dataframe = pd.DataFrame([])
for csv in csvList:
data=pd.read_csv(self.folderPath+'/'+csv,engine='python')
if(len(data)>self.date_duration):
data=data[-self.date_duration-1:-1]
data=data.reset_index()
data=data['시가']
dataframe=dataframe.append(data,ignore_index=True)
dataT=np.array(dataframe).T
self.scaler.fit(dataT)
dataT=self.scaler.transform(dataT)
dataT=dataT.T
dataframe=pd.DataFrame(dataT)
dataframe=dataframe.transpose()
print('StockPrice shape: ',dataframe.shape)
return dataframe
###EQUATION: ############### y=-7.018333333333211e-06x+10254.897027666497 #############
##############################OR y=4.120264489871436e-06x + -5250.689416195902#########
# n_samples = 30
epochs = 200
training_rate = 0.01
n_samples = 1570
start_date = '2013-12-01'
end_date = '2020-03-01'
# Set the ticker
ticker = 'GOOGL'
# Get the data
data = data.get_data_yahoo(ticker, start_date, end_date)
data.reset_index(level=0, inplace=True)
#data.to_csv("stocks.csv")
data = data.iloc[::-1]
# x= np.linspace(0,1570,100)
# y = 4.120264489871436e-06*x + -5250.689416195902
# plt.plot(x,y)
plt.show()
print(data.shape)
# test_x = np.linspace(0,35, n_samples)
# test_y = 5* test_x + 5 * np.random.randn(n_samples)
# plt.plot(test_x, test_y, 'o')
data['Timestamp'] = data['Date'].apply(lambda date: time.mktime(
data.head()
# create a new dataframe with 'date' column as index
new = data.set_index('date')
# use pandas pivot function to sort adj_close by tickers
clean_data = new.pivot(columns='ticker')
# check the head of the output
clean_data.head()
import quandl
import pandas as pd
import datetime
from datetime import timedelta
from pandas import ExcelWriter
quandl.ApiConfig.api_key = 'NL6weZFnRNw5q_L5-Nve'
tickers = ['BA', 'MSFT', 'ACN']
data = quandl.get_table('WIKI/PRICES',
ticker=tickers,
qopts={'columns': ['ticker', 'date', 'open', 'close']},
date={
'gte': '2000-01-01',
'lte': '2018-04-16'
},
paginate=True)
data.reset_index()
data.set_index('date')