☰
›

zerodha API and PCR

ashokbinra
ashokbinra in API clients
hi whats wrong with my coding , it runs successfully but taking too much time to finish the task , it is taking 40 min to 2 hrs , i have laptop with configuration of i7 and 8gb ram
THE FOLLOWING ARE MY CODE
rom nsepy import get_history

import pandas as pd

import numpy as np
import kiteconnect
from datetime import datetime, date, timedelta
from kiteconnect import KiteConnect

def write_log(text): f = open(r'log.txt', 'a') f.write("{}\n".format(text))

class ExcelSheetUpdate(object):

def __init__(self, **kwargs):

self.start_date = self.end_date = kwargs.get('date')
self.csv_file_name = file_name

# Tickers List
self.tickers = [
"ACC","ADANIENT","ADANIPORTS","AMARAJABAT","AMBUJACEM","APOLLOHOSP","APOLLOTYRE","ASHOKLEY","ASIANPAINT","AUROPHARMA",
"AXISBANK","BAJAJ-AUTO","BAJFINANCE","BAJAJFINSV","BALKRISIND","BANDHANBNK","BANKBARODA","BATAINDIA","BERGEPAINT","BEL",
"BHARATFORG","BPCL","BHARTIARTL","INFRATEL","BHEL","BIOCON","BOSCHLTD","BRITANNIA","CADILAHC","CANBK","CENTURYTEX",
"CHOLAFIN","CIPLA","COALINDIA","COLPAL","CONCOR","CUMMINSIND","DABUR","DIVISLAB","DLF","DRREDDY","EICHERMOT","EQUITAS",
"ESCORTS","EXIDEIND","FEDERALBNK","GAIL","GLENMARK","GMRINFRA","GODREJCP","GODREJPROP","GRASIM","HAVELLS","HCLTECH",
"HDFCBANK","HDFC","HDFCLIFE","HEROMOTOCO","HINDALCO","HINDPETRO","HINDUNILVR","ICICIBANK","ICICIPRULI","NAUKRI",
"IDEA","IDFCFIRSTB","IBULHSGFIN","IOC","IGL","INDUSINDBK","INFY","INDIGO","ITC","JINDALSTEL","JSWSTEEL","JUBLFOOD",
"JUSTDIAL","KOTAKBANK","L&TFH","LT","LICHSGFIN","LUPIN","M&MFIN","MGL","M&M","MANAPPURAM","MARICO","MARUTI","MFSL",
"MINDTREE","MOTHERSUMI","MRF","MUTHOOTFIN","NATIONALUM","NCC","NESTLEIND","COFORGE","NMDC","NTPC","ONGC","PAGEIND",
"PETRONET","PIDILITIND","PEL","PFC","POWERGRID","PNB","PVR","RBLBANK","RELIANCE","RECLTD","SHREECEM","SRTRANSFIN",
"SIEMENS","SRF","SBIN","SBILIFE","SAIL","SUNPHARMA","SUNTV","TATACHEM","TCS","TATACONSUM","TATAMOTORS","TATAPOWER",
"TATASTEEL","TECHM","RAMCOCEM","TITAN","TORNTPHARM","TORNTPOWER","TVSMOTOR","UJJIVAN","ULTRACEMCO","UBL","MCDOWELL-N",
"UPL","VEDL","VOLTAS","WIPRO","ZEEL","NIFTY","BANKNIFTY"
]

cash_df = pd.DataFrame() # this variable will contain fetched cash data

print('CASH DATA EXTRACTION BEGINS:', datetime.today().time())
write_log(text=['CASH DATA EXTRACTION BEGINS:', datetime.today().time()])

# Connecting To Zerodha API (Kite)
self.kite = KiteConnect(api_key='crpko3b6x4nw47ym')
self.kite.set_access_token(kite_token)

# Fetching Derivatives data from API
derivative_data = self.kite.instruments(exchange='NFO')
derivative_data = pd.DataFrame(derivative_data)
derivative_data.expiry = derivative_data.expiry.astype('datetime64[ns]').dt.date

# Slicing options data from derivative data
self.pc_data = derivative_data[derivative_data['instrument_type'] != 'FUT']
self.pc_data.reset_index(drop=True, inplace=True)

pcr = [] # this variable will contain PCR

print('PCR CALCULATION BEGINS:', datetime.today().time())
write_log(text=['PCR CALCULATION BEGINS:', datetime.today().time()])

# this loop will calculate PCR
for j in self.tickers:

print(j, datetime.today().time())
write_log(text=[j, datetime.today().time()])


finally:

if chance < 5:
if ans > 2.5: # this loop is for replacing PCR value, when we get PCR greater than 2.5
ans = 2.5

if len(expiry) == 1 and index == 0:
ratio = {'Symbol': j, 'PCR': ans, 'Weekly PCR': np.nan}
pcr.append(ratio)
elif len(expiry) == 2 and index == 0:
ratio = {'Symbol': j, 'PCR': ans, 'Weekly PCR': np.nan}
pcr.append(ratio)
elif len(expiry) == 2 and index == 1:
pcr[-1]['Weekly PCR'] = ans

pcr_df = pd.DataFrame(pcr) # this dataframe will now contain PCR of different tickers

print('FUTURE OI CALCULATION BEGINS:', datetime.today().time())
write_log(text=['FUTURE OI CALCULATION BEGINS:', datetime.today().time()])

# Slicing futures data from derivative data
futures_expiry_date = self.get_expiry(self.start_date, symbol='RELIANCE', w_m='MONTHLY')
future_data = derivative_data[derivative_data['instrument_type'] == 'FUT']
future_data = future_data[future_data['expiry'] == futures_expiry_date]

oi_data = [] # this variable will contain OI



print('CREATION OF FINAL DF FOR CSV FILE BEGINS:', datetime.today().time())
write_log(text=['CREATION OF FINAL DF FOR CSV FILE BEGINS:', datetime.today().time()])

self.final_data = pd.merge(cash_df, pcr_df, how='outer')
self.final_data = pd.merge(self.final_data, oi_df, how='outer')
self.final_data['Date'] = self.start_date.strftime("%d-%m-%Y")
self.final_data = self.final_data[['Date', 'Symbol', 'Open', 'High', 'Low', 'Close', 'Volume', '%Deliverble', 'OI', 'PCR', 'Weekly PCR']]

# Appending extracted data into exisiting file
self.final_data.to_csv(self.csv_file_name, mode='a', index=False, header=None)

print('DATA APPENDED SUCCESSFULLY IN CSV FILE', datetime.today().time())
write_log(text=['DATA APPENDED SUCCESSFULLY IN CSV FILE', datetime.today().time()])



# this function will fetch OHLC, volume, %Deliverable
def cash_data(self, symbol):

if symbol == 'NIFTY' or symbol == 'BANKNIFTY':
df = get_history(symbol=symbol, start=self.start_date, end=self.end_date, index=True)
df = df[['Open', 'High', 'Low', 'Close', 'Volume']]
df['Symbol'] = symbol
df['%Deliverble'] = np.nan
else:
df = get_history(symbol=symbol, start=self.start_date, end=self.end_date, index=False)
df = df[['Symbol', 'Open', 'High', 'Low', 'Close', 'Volume', '%Deliverble']]

return df

# this function will fetch options data
def options_data(self, symbol, expiry):

df = self.pc_data[self.pc_data['expiry'] == expiry]

df = df[df['name'] == symbol]
df.reset_index(drop=True, inplace=True)

call_inst_token = df[df['instrument_type'] == 'CE']['instrument_token'].tolist()
put_inst_token = df[df['instrument_type'] == 'PE']['instrument_token'].tolist()

total_call_oi = 0
for ele1 in call_inst_token:
res1 = self.kite.historical_data(ele1, self.start_date, self.end_date, 'day', continuous=False, oi=True)

if len(res1) != 0:
total_call_oi += res1[0]['oi']

total_put_oi = 0
for ele2 in put_inst_token:
res2 = self.kite.historical_data(ele2, self.start_date, self.end_date, 'day', continuous=False, oi=True)
if len(res2) != 0:
total_put_oi += res2[0]['oi']

try:
ratio = total_put_oi / total_call_oi
except ZeroDivisionError:
ratio = 0

return ratio
Tagged:
Sign In or Register to comment.