Hi all

Just after I posted that question, I found a better way by using getdatabyname(). See below.
However, I still want to clarify my understanding that the Strategy class needs to identify which datas[] to use. Right?

class MyStrategy(bt.Strategy):
    params=(('data_name', ''),)

    def __init__(self):
        pass

    def next(self):

        # Find the data feed we need.
        inst = self.params.data_name
        inst_data = self.getdatabyname(inst)

        # Print what was selected.
        dt = inst_data.datetime.date(0)
        dn = inst_data._name
        print(f'{dt.isoformat():s}: instrument is {inst:s}')
        print(f'            inst_data is {dn:s}')

Tony

@vladisld Thanks. That makes sense. Much cleaner to do it in __init__().

Hi all

Just after I posted that question, I found a better way by using getdatabyname(). See below.
However, I still want to clarify my understanding that the Strategy class needs to identify which datas[] to use. Right?

class MyStrategy(bt.Strategy):
    params=(('data_name', ''),)

    def __init__(self):
        pass

    def next(self):

        # Find the data feed we need.
        inst = self.params.data_name
        inst_data = self.getdatabyname(inst)

        # Print what was selected.
        dt = inst_data.datetime.date(0)
        dn = inst_data._name
        print(f'{dt.isoformat():s}: instrument is {inst:s}')
        print(f'            inst_data is {dn:s}')

Tony

Hi all

Is there a convenient way to associate the Strategy class with a particular data feed?

I'm trying to understand how to implement a portfolio system where each instrument uses a slightly different variation of the trading rules.
I have read these to blog posts (very good), but I want to check my understanding before going much further.
https://www.backtrader.com/blog/posts/2017-04-09-multi-example/multi-example/
https://backtest-rookies.com/2017/08/22/backtrader-multiple-data-feeds-indicators/

I think the strategy code needs to identify which data feed to use and this example shows how I do it, but is there a better way already provided?

I'm loving Backtrader but still learning.

I'd appreciate any guidance.

Tony

from datetime import datetime
import backtrader as bt

class MyStrategy(bt.Strategy):
    params=(('data_name', ''),)

    def __init__(self):
        pass

    def next(self):
        # Find the data feed we need.
        data_idx = 0
        for i, d in enumerate(self.datas):
            if self.datas[i]._name == self.params.data_name:
                data_idx = i
                
        # From here, use only datas[data_idx]...

        # Print what was selected.
        dt = self.datas[data_idx].datetime.date(0)
        dn = self.datas[data_idx]._name
        inst = self.params.data_name
        print(f'{dt.isoformat():s}: instrument is {inst:s}')
        print(f'            datas[{data_idx:d}] is {dn:s}')


if __name__ == '__main__':

    instruments = ['NQ=F', 'GC=F']

    from_date = datetime(2019, 12, 1)
    to_date = datetime(2020, 1, 1)

    # Create the brain.
    cerebro = bt.Cerebro()

    for inst in instruments:
        # Add the instrument data feeds.
        data_file = '../../Data/Yahoo Data/' + inst + '.csv'
        data = bt.feeds.YahooFinanceCSVData(dataname=data_file, name=inst,
                                            fromdate=from_date, todate=to_date)
        cerebro.adddata(data)

        # Add one strategy for each instrument (data feed).
        strat = cerebro.addstrategy(MyStrategy, data_name=inst)
        print('Strategy added for instrument ' + inst)

    cerebro.run()