algo-trading-zerodha's Introduction

Algo-Trading

This repository contains the code snippets used for trading with Zerodha KiteConnect API using python

A Hub of necessary python functions for algorithmic trading

Key Features • How To Use • Download • Credits • Related • License

How To Use

To use the functions, you'll need to write the following code beforehand.

from kiteconnect import KiteConnect
import logging
import os
import datetime as dt
import pandas as pd

cwd = os.chdir("Current Working Directory")

# generate trading sesion
# Please go to Gen Access folder to get these two files as input
access_token = open("access_token.txt", 'r').read()
key_secret = open("api_key.txt", 'r').read().split()
kite = KiteConnect(api_key=key_secret[0])
kite.set_access_token(access_token)
                 

# get dump of all NSE instruments
instrument_dump = kite.instruments("NSE")
instrument_df = pd.DataFrame(instrument_dump)
instrument_df.to_csv("NSE_Instruments.csv", index = False)

def instrumentLookup(instrument_df, symbol):
    try:
        return instrument_df[instrument_df['tradingsymbol']==symbol]['instrument_token'].values[0]
    except:
        return -1
    
def fetchOHLC(ticker, interval, from_date, to_date):
    """extracts historical data and outputs in the form of dataframe"""
    # ticker is the NSE instrument you want to use
    # interval is the candlestick time frame for OHLC data
    # The dates should be in this form - "day-month-Year"
    instrument = instrumentLookup(instrument_df, ticker)
    from_date = dt.datetime.strptime(from_date, '%d-%m-%Y')
    to_date = dt.datetime.strptime(to_date, '%d-%m-%Y')
    data = pd.DataFrame(columns=['date', 'open', 'high', 'low', 'close', 'volume'])
    while True:
        if from_date.date() >= dt.date.today()-dt.timedelta(100):        
            data = data.append(pd.DataFrame(kite.historical_data(instrument, from_date, to_date, interval)))
            break
        else:
            to_date = from_date + dt.timedelta(100)
            data = data.append(pd.DataFrame(kite.historical_data(instrument, from_date, to_date, interval)))
            from_date = to_date
    data.set_index("date", inplace=True)
    return data

Recommend Projects