AttributeError: 'numpy.int64' object has no attribute 'lower' when trying to run Cerebro after importing data from Pandas

[_will_derness]

Hello,

Kind of new to all this so apologies if I am missing something basic.

I have tried to search through all the information about using pd dataframes with backtrader and have had a look through other people's posts on here and one was kind of similar i think but i couldn't see a solution and for some reason couldn't reply to it.

When i add the data i don't get an error but when i run i get a long error ending in this "AttributeError: 'numpy.int64' object has no attribute 'lower'".

Here is my code in case that helps:

import matplotlib.pyplot as plt
import numpy as np
import pandas as pd
import scipy.stats as st

import datetime
import requests
import backtrader as bt
import backtrader.feeds as btfeeds
from strategies import *

cerebro = bt.Cerebro()

#Add data to Cerebro
def get_bitfinex_asset(asset, ts_ms_start, ts_ms_end):
    url = 'https://api.bitfinex.com/v2/candles/trade:1D:t' + asset + '/hist'
    params = { 'start': ts_ms_start, 'end': ts_ms_end, 'sort': 1}
    r = requests.get(url, params = params)
    data_dl = r.json()
    return pd.DataFrame(data_dl)

start_date = 1514768400000 # 1 January 2018, 00:00:00
end_date = 1527811199000   # 31 May 2018, 23:59:59
global_asset = 'ETHBTC'

print('Downloading ' + global_asset)
dataframe = get_bitfinex_asset(asset = global_asset, ts_ms_start = start_date, ts_ms_end = end_date)
print(dataframe)

data = bt.feeds.PandasData(dataname=dataframe)

cerebro.adddata(data)

cerebro.addstrategy(PrintClose)

cerebro.run()

If i comment out the cerebro.run() line I dont get the error, and the rest of the console output looks like this:

Any ideas about how I might be able to get backtrader to properly process my data?

[ab_trader]

@_will_derness you may want to read docs one more time to learn how data feeds are organized, and how pandas data frame should be formatted to be used with bt.

https://www.backtrader.com/docu/datafeed/
https://www.backtrader.com/docu/pandas-datafeed/pandas-datafeed/

[vladisld]

It seems a very similar question was discussed already not so long ago. Take a look:

https://community.backtrader.com/topic/3411/data-feed-format

[run-out]

I could be off, but check your dataframe columns. "lower" sounds suspicious. Should be 'low' me thinks.

[run-out]

@run-out said in AttributeError: 'numpy.int64' object has no attribute 'lower' when trying to run Cerebro after importing data from Pandas:

I could be off, but check your dataframe columns. "lower" sounds suspicious. Should be 'low' me thinks.

I was off. This error is most likely due to you haveing no headers in your data set. Here's the error I get when I try to run a backtest with no headers.

File "/home/runout/a_bt_env/backtrader/backtrader/feed.py", line 203, in _start
    self.start()
  File "/home/runout/a_bt_env/backtrader/backtrader/feeds/pandafeed.py", line 212, in start
    colnames = [x.lower() for x in self.p.dataname.columns.values]
  File "/home/runout/a_bt_env/backtrader/backtrader/feeds/pandafeed.py", line 212, in <listcomp>
    colnames = [x.lower() for x in self.p.dataname.columns.values]
AttributeError: 'numpy.int64' object has no attribute 'lower'

This is a known Issue #43 which I dropped the ball on a couple of months ago. I will pick it up now.

In the mean time, if you can add headers ['open', 'high', 'low', 'close', 'volume'] headers to your data you should be ok for now.