How to reset the tradehistory

chhrissi2909

Hello,

In my code I loop through all my symbols. In detail I backtest a strategy with cerebro on a symbol and if it is ready I loop through the next stock.

For analyzing the results I use the "class trade_list(bt.Analyzer):" (https://community.backtrader.com/topic/1274/closed-trade-list-including-mfe-mae-analyzer)

If I just loop through 1 Stock there is no problem and the code runs well. But if I loop through 2 or more Stocks, than this error message shows:

"TypeError: 'list' object is not callable"

I think the problem is with this line in my cerebro plot:

strats = cerebro.run(tradehistory=True)

I think I have to reset the trade history, if the next Stock is looping. (If I print my Sticks than I can see, that the first stock loops through and as soo as the second Stock comes, the code stops)

Is there a opportunitie to reset the tradehystory, if my loop starts with the next Stock. For example like this:

if i == 0:
set_tradehystory = False 
else:
set_tradehistory=true

thank you for your help :=)

best regards
Christian

vladisld

It would be great if you could share the full code and the full backtrace of the exception you've got.

AFAIU the trade history is kept per trade, where the trades are accumulated per strategy. The strategy instances are kept per cerebro instance. If you are using a new cerebro instance per symbol - which I understand is what happening - no trading history need to be cleaned.

However until we see the whole source code - it is hard to tell.

chhrissi2909

Hello vladisld,

I can't post my whole code, but I try to give you an overview about my logic:

class trade_list(bt.Analyzer):
      # --> pasted code of the link in the problem description
class ZigZagStrategy(bt.Strategy):
     # --> my Strategy

if __name__ == '__main__':
    for i in range(len(alleSymbole)):
        print("(", alleSymbole[i][0],")", ".... process complete ....   ",round(i/len(alleSymbole)*100,2),"%")
        cerebro = bt.Cerebro()
        datapath = DATA_LOCATION+alleSymbole[i][0]+".csv"
        data_df = pd.read_csv(datapath, 
                            index_col=0, 
                            usecols=["date", "open","high", "low", "close"])
        data_df.index = pd.to_datetime(data_df.index)
        feed = bt.feeds.PandasData(dataname=data_df,name=alleSymbole[i][0]) 
        cerebro.adddata(feed)
        cerebro.broker.setcash(STARTING_CAP)
     
        cerebro.broker.setcommission(commission=COMMISSION, leverage=2)
        cerebro.addanalyzer(TotalTrades, _name='total_trades')

        strats = cerebro.run(tradehistory=True)
            # get analyzers data
        trade_list = strats[0].analyzers.trade_list.get_analysis()
        trade_list_tabuliert = tabulate(trade_list, headers="keys")
        trade_list_pandas = pd.DataFrame(trade_list)


        
        # 1. calculating profit factor 
        winners = 0
        winner_counter = 0
        loosers = 0
        looser_counter = 0
        for q in range(len(trade_list_pandas["pnl"])):
            if trade_list_pandas["pnl"][q] >= 0:
                winners += trade_list_pandas["pnl"][q]
                winner_counter += 1
            else:
                loosers += trade_list_pandas["pnl"][q]
                looser_counter += 1
        winners = winners / winner_counter
        loosers = (loosers / looser_counter) * (-1)
        profit_factor[i] = round(winners/loosers,2)

        for l in range(len(trade_list_pandas["cumpnl"])):
            Stock_Equity[l][i] = round(trade_list_pandas["cumpnl"][l],2)


   symbole = np.empty(len(alleSymbole),dtype=str)
   for i in range(len(alleSymbole)):
   symbole[i] = alleSymbole[i][0]

    pandas = pd.DataFrame(Stock_Equity.tolist(),columns=symbole)
    PATHi = "/Users/MYPATH.xlsx"
    pandas.to_excel(PATHi, index=True)

My Error message is here:

File "/Users/PATH", line 484, in <module>
  strats = cerebro.run(tradehistory=True)
File "/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/backtrader/cerebro.py", line 1127, in run
  runstrat = self.runstrategies(iterstrat)
File "/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/backtrader/cerebro.py", line 1257, in runstrategies
  strat._addanalyzer(ancls, *anargs, **ankwargs)
Fi**le "/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/backtrader/strategy.py", line 247, in _addanalyzer
  analyzer = ancls(*anargs, **ankwargs)
TypeError: 'list' object is not callable

vladisld

@chhrissi2909 said in How to reset the tradehistory:

cerebro.addanalyzer(TotalTrades, _name='total_trades')

How the TotalTrades class is defined?

chhrissi2909

@vladisld said in How to reset the tradehistory:

TotalTrades

class TotalTrades(bt.Analyzer):
    def __init__(self):
        self.tradeanalyzer = bt.analyzers.TradeAnalyzer()
    def get_analysis(self):
        tradeanalyzer = self.tradeanalyzer.get_analysis()
        try:
            tot_trades = tradeanalyzer.total.closed
        except:
            tot_trades = 0
        return tot_trades

chhrissi2909

@chhrissi2909
and I run these analyzers like this:

results = cerebro.run()
total_trades[i] = results[0].analyzers.total_trades.get_analysis()

so I use

strats = cerebro.run(tradehistory=True)

for the pasted class trade_list
and

results = cerebro.run()

for my own analyzers..

could this be the problem?

vladisld

@chhrissi2909 said in How to reset the tradehistory:

analyzer = ancls(*anargs, **ankwargs)
TypeError: 'list' object is not callable

Could be a stupid question: are you sure there are no other variable called TotalTrades in your code ? ( according to the error message a simple list instance was passed to the addanalyzer method instead of class name )

chhrissi2909

@vladisld said in How to reset the tradehistory:

TotalTrades

Hello vladisld,
no that is not a stupid question, but in my whole script the "word" "TotalTrades" only appears twice. The first in the class name:

class TotalTrades(bt.Analyzer):

and secondly in the addanalyzer:

cerebro.addanalyzer(TotalTrades, _name='total_trades')

vladisld

Sorry, wasn't able to repro given the provided code sample. I've filled the missing part of code by some common sense code - and it was running without raising any exception.

Since no full original code is available I can only assume the problem is somewhere in your code and not in Backtrader's code.

Would suggest to try to debug it yourself by breaking at the cerebro.runstrategies and see what has been passed to the strat._addanalyzer call.