Anyone use backtrader to do live trading on Bitcoin exchange?

Nguyễn Tài Nguyên

from __future__ import (absolute_import, division, print_function,
						unicode_literals)

import sys

import backtrader as bt


class TestStrategy(bt.Strategy):
	def notify_data(self, data, status, *args, **kwargs):
		print('*' * 5, 'DATA NOTIF:', data._getstatusname(status))

	def next(self):
		print('*' * 5, 'NEXT:', bt.num2date(self.data0.datetime[0]), self.data0._name, self.data0.open[0],
			  bt.TimeFrame.getname(self.data0._timeframe), len(self.data0))


def runstrategy(argv):
	# Create a cerebro
	cerebro = bt.Cerebro()

	data = bt.feeds.CCXT(exchange='gdax', symbol='BTC/USD', timeframe=bt.TimeFrame.Ticks, compression=1)
	cerebro.resampledata(data, timeframe=bt.TimeFrame.Seconds)
	# cerebro.adddata(data)

	# Add the strategy
	cerebro.addstrategy(TestStrategy)

	# Run the strategy
	cerebro.run()


if __name__ == '__main__':
	sys.exit(runstrategy(sys.argv))

@Ed-Bartosh . When I run this above code, it shows an error. How to fix it ?

  File "C:/Users/PC/PycharmProjects/trading/scr/test.py", line 34, in <module>
	sys.exit(runstrategy(sys.argv))
  File "C:/Users/PC/PycharmProjects/trading/scr/test.py", line 30, in runstrategy
	cerebro.run()
  File "C:\Users\PC\PycharmProjects\trading\venv\lib\site-packages\backtrader\cerebro.py", line 1127, in run
	runstrat = self.runstrategies(iterstrat)
  File "C:\Users\PC\PycharmProjects\trading\venv\lib\site-packages\backtrader\cerebro.py", line 1295, in runstrategies
	self._runnext(runstrats)
  File "C:\Users\PC\PycharmProjects\trading\venv\lib\site-packages\backtrader\cerebro.py", line 1538, in _runnext
	drets.append(d.next(ticks=False))
  File "C:\Users\PC\PycharmProjects\trading\venv\lib\site-packages\backtrader\feed.py", line 404, in next
	ret = self.load()
  File "C:\Users\PC\PycharmProjects\trading\venv\lib\site-packages\backtrader\feed.py", line 476, in load
	_loadret = self._load()
  File "C:\Users\PC\PycharmProjects\trading\venv\lib\site-packages\backtrader\feeds\ccxt.py", line 91, in _load
	self._fetch_ohlcv()
  File "C:\Users\PC\PycharmProjects\trading\venv\lib\site-packages\backtrader\feeds\ccxt.py", line 110, in _fetch_ohlcv
	granularity = self.store.get_granularity(self._timeframe, self._compression)
  File "C:\Users\PC\PycharmProjects\trading\venv\lib\site-packages\backtrader\stores\ccxtstore.py", line 73, in get_granularity
	(bt.TimeFrame.getname(timeframe), compression))
  File "C:\Users\PC\PycharmProjects\trading\venv\lib\site-packages\backtrader\dataseries.py", line 45, in getname
	if compression > 1 or tname == cls.Names[-1]:
TypeError: unorderable types: NoneType() > int()

Ed Bartosh

@nguyễn-tài-nguyên said in Anyone use backtrader to do live trading on Bitcoin exchange?:

cerebro.resampledata(data, timeframe=bt.TimeFrame.Seconds)

You can try to pass compression to resampledata. This should theoreticall fix this.

cerebro.resampledata(data,
                     timeframe=bt.TimeFrame.Seconds,
                     compression=1)

Ed Bartosh

@søren-pallesen said in Anyone use backtrader to do live trading on Bitcoin exchange?:

@Nguyễn-Tài-Nguyên I have the same issue. Its a quedtion for @Ed-Bartosh that I hope can spare some time soon to help us out.

Sorry for this delay. Vacation time.

Will look at this today or tomorrow.

Søren Pallesen

Thx Ed - really appreciate that :-) ...hope you enjoyed the time off ...

Nguyễn Tài Nguyên

edited this post

Ed Bartosh

@søren-pallesen said in Anyone use backtrader to do live trading on Bitcoin exchange?:

Thx Ed - really appreciate that :-) ...hope you enjoyed the time off ...

I started to look at it today. I can see the issue with bitmex, but not with gdax. Can you explain what kind of output you expect? It would be great to show correct behaviour using another data feed (ib, oanda, etc).

Søren Pallesen

@Ed-Bartosh thx for looking at this.

I expect the real time feed to continue so that i can trade against it live.

Ed Bartosh

@søren-pallesen said in Anyone use backtrader to do live trading on Bitcoin exchange?:

@Ed-Bartosh thx for looking at this.

I expect the real time feed to continue so that i can trade against it live.

Can you show an example of this using ib or oanda data feed?

Søren Pallesen

@Ed-Bartosh That would take some time to setup needing accounts for API etc. and before doing that please answer the if the following use cases are supported NOW by your implementation of CCXT integration:

Use case 1: Live data feed from a CCXT exchange AND trading against a Backtrader broker i.e. simulated trading (I have provided my example of this in previous posts but the live feed stops)
Use case 2: Live data feed from a CCXT exchange AND live trading against a CCXT broker i.e. real live trading

If you have either Use case 1 or 2 working could you please share a simple code examples? Also which exchanges have you tested and feel works? (for example you say that Gdax is working for you - please share example and a description of the behavior you are getting)

The main reason that I an picking Bitmex as my test exchange is that 1) the API is stable and robust and 2) the have a reliable sandbox (https://testnet.bitmex.com/) which makes testing safe.

My current objectives is to first get Use case 1 working with any exchange then in phase 2 get Use case 2 working on https://testnet.bitmex.com/.

tornadoatc

@Ed-Bartosh Hi, I am interested in your integration efforts with CCXT / backtrader / zipline.. I am wondering if you are familiar with the enigma catalyst effort (https://github.com/enigmampc/catalyst/tree/0.4.3) ?? As of this version, they have integrated the ccxt api. May this open source effort be of assistance?

Can you comment on what you strengths and weaknesses between catalyst and backtrader ? I am looking to jump in but wondering which tool to focus on.

Ed Bartosh

@tornadoatc said in Anyone use backtrader to do live trading on Bitcoin exchange?:

@Ed-Bartosh Hi, I am interested in your integration efforts with CCXT / backtrader / zipline.. I am wondering if you are familiar with the enigma catalyst effort (https://github.com/enigmampc/catalyst/tree/0.4.3) ?? As of this version, they have integrated the ccxt api. May this open source effort be of assistance?

Can you comment on what you strengths and weaknesses between catalyst and backtrader ? I am looking to jump in but wondering which tool to focus on.

Hi,

Thank you for the info. I didn't know about enigma catalyst before.
As enigma is based on zipline I'd suggest you to look at zipline-live project. As one of its owners I can ensure they'll be happy to collaborate.

Regards,
Ed

Ed Bartosh

@søren-pallesen said in Anyone use backtrader to do live trading on Bitcoin exchange?:

@Ed-Bartosh That would take some time to setup needing accounts for API etc. and before doing that please answer the if the following use cases are supported NOW by your implementation of CCXT integration

Use case 1: Live data feed from a CCXT exchange AND trading against a Backtrader broker i.e. simulated trading (I have provided my example of this in previous posts but the live feed stops)

This should be fixed by this commit: https://github.com/bartosh/backtrader/commit/d6a82e3ca3b85239770b368c4814b1f138195036

here is a test script:

#!/usr/bin/env python
# -*- coding: utf-8; py-indent-offset:4 -*-
###############################################################################
#
# Copyright (C) 2017,2018 Ed Bartosh <bartosh@gmail.com>
#
# This program is free software: you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation, either version 3 of the License, or
# (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program.  If not, see <http://www.gnu.org/licenses/>.
#
###############################################################################
from __future__ import (absolute_import, division, print_function,
                        unicode_literals)

import sys

import backtrader as bt

class TestStrategy(bt.Strategy):
    def next(self):
        for data in self.datas:
            print('*' * 5, 'NEXT:', bt.num2date(data.datetime[0]), data._name, data.open[0],
                  data.high[0], data.low[0], data.close[0], data.volume[0],
                  bt.TimeFrame.getname(data._timeframe), len(data))
            if not self.getposition(data):
                self.buy(data, exectype=bt.Order.Limit, size=10, price=data.close[0])

    def notify_order(self, order):
        print('*' * 5, "NOTIFY ORDER", order)

def runstrategy(argv):
    # Create a cerebro
    cerebro = bt.Cerebro()

    # Create data feeds
    data_ticks = bt.feeds.CCXT(exchange='gdax', symbol='BTC/USD', name="btc_usd_tick",
                               timeframe=bt.TimeFrame.Ticks, compression=1)
    cerebro.adddata(data_ticks)

    # Add the strategy
    cerebro.addstrategy(TestStrategy)

    # Run the strategy
    cerebro.run()

if __name__ == '__main__':
    sys.exit(runstrategy(sys.argv))

and its output:

(backtrader-ccxt) backtrader-ccxt (ccxt) $ ./bt-ccxt-no-broker.py 
***** NEXT: 2018-01-07 23:27:02.790004 btc_usd_tick 16319.71 16319.71 16319.71 16319.71 0.42092738 Tick 1
***** NOTIFY ORDER Ref: 1
OrdType: 0
OrdType: Buy
Status: 1
Status: Submitted
Size: 10
Price: 16319.71
Price Limit: None
TrailAmount: None
TrailPercent: None
ExecType: 2
ExecType: Limit
CommInfo: None
End of Session: 736702.0
Info: AutoOrderedDict()
Broker: None
Alive: True
***** NOTIFY ORDER Ref: 1
OrdType: 0
OrdType: Buy
Status: 7
Status: Margin
Size: 10
Price: 16319.71
Price Limit: None
TrailAmount: None
TrailPercent: None
ExecType: 2
ExecType: Limit
CommInfo: None
End of Session: 736702.0
Info: AutoOrderedDict()
Broker: None
Alive: False
***** NEXT: 2018-01-07 23:27:06.859002 btc_usd_tick 16322.0 16322.0 16322.0 16322.0 0.15175919 Tick 2
***** NOTIFY ORDER Ref: 2
OrdType: 0
OrdType: Buy
Status: 1
Status: Submitted
Size: 10
Price: 16322.0
Price Limit: None
TrailAmount: None
TrailPercent: None
ExecType: 2
ExecType: Limit
CommInfo: None
End of Session: 736702.0
Info: AutoOrderedDict()
Broker: None
Alive: True
***** NOTIFY ORDER Ref: 2
OrdType: 0
OrdType: Buy
Status: 7
Status: Margin
Size: 10
Price: 16322.0
Price Limit: None
TrailAmount: None
TrailPercent: None
ExecType: 2
ExecType: Limit
CommInfo: None
End of Session: 736702.0
Info: AutoOrderedDict()
Broker: None
Alive: False
***** NEXT: 2018-01-07 23:27:09.158004 btc_usd_tick 16323.79 16323.79 16323.79 16323.79 0.0035 Tick 3
***** NOTIFY ORDER Ref: 3
OrdType: 0
OrdType: Buy
Status: 1
Status: Submitted
Size: 10
Price: 16323.79
Price Limit: None
TrailAmount: None
TrailPercent: None
ExecType: 2
ExecType: Limit
CommInfo: None
End of Session: 736702.0
Info: AutoOrderedDict()
Broker: None
Alive: True
***** NOTIFY ORDER Ref: 3
OrdType: 0
OrdType: Buy
Status: 7
Status: Margin
Size: 10
Price: 16323.79
Price Limit: None
TrailAmount: None
TrailPercent: None
ExecType: 2
ExecType: Limit
CommInfo: None
End of Session: 736702.0
Info: AutoOrderedDict()
Broker: None
Alive: False
***** NEXT: 2018-01-07 23:27:18.672004 btc_usd_tick 16332.04 16332.04 16332.04 16332.04 0.011213 Tick 4
***** NOTIFY ORDER Ref: 4
OrdType: 0
OrdType: Buy
Status: 1
Status: Submitted
Size: 10
Price: 16332.04
Price Limit: None
TrailAmount: None
TrailPercent: None
ExecType: 2
ExecType: Limit
CommInfo: None
End of Session: 736702.0
Info: AutoOrderedDict()
Broker: None
Alive: True
***** NOTIFY ORDER Ref: 4
OrdType: 0
OrdType: Buy
Status: 7
Status: Margin
Size: 10
Price: 16332.04
Price Limit: None
TrailAmount: None
TrailPercent: None
ExecType: 2
ExecType: Limit
CommInfo: None
End of Session: 736702.0
Info: AutoOrderedDict()
Broker: None
Alive: False
***** NEXT: 2018-01-07 23:27:18.672004 btc_usd_tick 16338.32 16338.32 16338.32 16338.32 1.19202592 Tick 5
***** NOTIFY ORDER Ref: 5
OrdType: 0
OrdType: Buy
Status: 1
Status: Submitted
Size: 10
Price: 16338.32
Price Limit: None
TrailAmount: None
TrailPercent: None
ExecType: 2
ExecType: Limit
CommInfo: None
End of Session: 736702.0
Info: AutoOrderedDict()
Broker: None
Alive: True
...

Use case 2: Live data feed from a CCXT exchange AND live trading against a CCXT broker i.e. real live trading

this is not working yet. It stops after getting historical price data and doesn't continue further. I'll look at it when I have time.

If you have either Use case 1 or 2 working could you please share a simple code examples? Also which exchanges have you tested and feel works? (for example you say that Gdax is working for you - please share example and a description of the behavior you are getting)

I only test it with gdax as a data feed and gemini as a broker.

The main reason that I an picking Bitmex as my test exchange is that 1) the API is stable and robust and 2) the have a reliable sandbox (https://testnet.bitmex.com/) which makes testing safe.

I'm going to fix bitmex data feed problem. It only produces one bar of historical data for some reason. It's item 1 in my todo list.

My current objectives is to first get Use case 1 working with any exchange then in phase 2 get Use case 2 working on https://testnet.bitmex.com/.

My current todo list has only 2 items:

fix bitmex historical data issue
make live trading after getting historical data working (issue number 2 in your list)

Let me know if you discover anything else. And thank you very much for the help!

Ed Bartosh

@søren-pallesen said in Anyone use backtrader to do live trading on Bitcoin exchange?:

That is one bar... i must be missing some thing basic - please help? :-)

This should be fixed in the latest code.
Here is the example script:

from __future__ import (absolute_import, division, print_function,
                        unicode_literals)

import sys
import time

from datetime import datetime, timedelta

import backtrader as bt

class TestStrategy(bt.Strategy):
    def next(self):
        for data in self.datas:
            print('*' * 5, 'NEXT:', bt.num2date(data.datetime[0]), data._name, data.open[0], data.high[0],
                  data.low[0], data.close[0], data.volume[0],
                  bt.TimeFrame.getname(data._timeframe), len(data))

    def notify_order(self, order):
        print('*' * 5, "NOTIFY ORDER", order)

def runstrategy(argv):
    # Create a cerebro
    cerebro = bt.Cerebro()

    # Create data feeds
    hist_start_date = datetime.utcnow() - timedelta(minutes=30)
    data_min = bt.feeds.CCXT(exchange="bitmex", symbol="BTC/USD", name="btc_usd_min",
                             timeframe=bt.TimeFrame.Minutes, fromdate=hist_start_date) #, historical=True)
    cerebro.adddata(data_min)

    # Add the strategy
    cerebro.addstrategy(TestStrategy)

    # Run the strategy
    cerebro.run()

if __name__ == '__main__':
    sys.exit(runstrategy(sys.argv))

And the output:

$ ./bt-ccxt-bitmex.py 
***** NEXT: 2018-01-08 00:38:00 btc_usd_min 16195.5 16200.5 16195.5 16200.5 84491.0 Minute 1
***** NEXT: 2018-01-08 00:39:00 btc_usd_min 16217.5 16217.0 16213.0 16213.0 22907.0 Minute 2
***** NEXT: 2018-01-08 00:40:00 btc_usd_min 16217.0 16217.0 16195.5 16195.5 99852.0 Minute 3

The issue with backtest on historical + live price data still remains unsolved. I probably messed something up in the feed code. Will try to fix it when I have time.

Any help is welcome as always.

Nguyễn Tài Nguyên

@ed-bartosh said in Anyone use backtrader to do live trading on Bitcoin exchange?:

@søren-pallesen said in Anyone use backtrader to do live trading on Bitcoin exchange?:

That is one bar... i must be missing some thing basic - please help? :-)

This should be fixed in the latest code.
Here is the example script:

from __future__ import (absolute_import, division, print_function,
                        unicode_literals)

import sys
import time

from datetime import datetime, timedelta

import backtrader as bt

class TestStrategy(bt.Strategy):
    def next(self):
        for data in self.datas:
            print('*' * 5, 'NEXT:', bt.num2date(data.datetime[0]), data._name, data.open[0], data.high[0],
                  data.low[0], data.close[0], data.volume[0],
                  bt.TimeFrame.getname(data._timeframe), len(data))

    def notify_order(self, order):
        print('*' * 5, "NOTIFY ORDER", order)

def runstrategy(argv):
    # Create a cerebro
    cerebro = bt.Cerebro()

    # Create data feeds
    hist_start_date = datetime.utcnow() - timedelta(minutes=30)
    data_min = bt.feeds.CCXT(exchange="bitmex", symbol="BTC/USD", name="btc_usd_min",
                             timeframe=bt.TimeFrame.Minutes, fromdate=hist_start_date) #, historical=True)
    cerebro.adddata(data_min)

    # Add the strategy
    cerebro.addstrategy(TestStrategy)

    # Run the strategy
    cerebro.run()

if __name__ == '__main__':
    sys.exit(runstrategy(sys.argv))

And the output:

$ ./bt-ccxt-bitmex.py 
***** NEXT: 2018-01-08 00:38:00 btc_usd_min 16195.5 16200.5 16195.5 16200.5 84491.0 Minute 1
***** NEXT: 2018-01-08 00:39:00 btc_usd_min 16217.5 16217.0 16213.0 16213.0 22907.0 Minute 2
***** NEXT: 2018-01-08 00:40:00 btc_usd_min 16217.0 16217.0 16195.5 16195.5 99852.0 Minute 3

The issue with backtest on historical + live price data still remains unsolved. I probably messed something up in the feed code. Will try to fix it when I have time.

Any help is welcome as always.
Thank you very much for your awesome work :)

Nguyễn Tài Nguyên

@ed-bartosh Hi, when I execute your code, it shows an error. How to fix it ? Thank you !

  File "C:/Users/PC/PycharmProjects/trading/scr/test.py", line 41, in <module>
	sys.exit(runstrategy(sys.argv))
  File "C:/Users/PC/PycharmProjects/trading/scr/test.py", line 37, in runstrategy
	cerebro.run()
  File "C:\Users\PC\PycharmProjects\trading\venv\lib\site-packages\backtrader\cerebro.py", line 1127, in run
	runstrat = self.runstrategies(iterstrat)
  File "C:\Users\PC\PycharmProjects\trading\venv\lib\site-packages\backtrader\cerebro.py", line 1295, in runstrategies
	self._runnext(runstrats)
  File "C:\Users\PC\PycharmProjects\trading\venv\lib\site-packages\backtrader\cerebro.py", line 1626, in _runnext
	strat._next()
  File "C:\Users\PC\PycharmProjects\trading\venv\lib\site-packages\backtrader\strategy.py", line 325, in _next
	super(Strategy, self)._next()
  File "C:\Users\PC\PycharmProjects\trading\venv\lib\site-packages\backtrader\lineiterator.py", line 268, in _next
	self.nextstart()  # only called for the 1st value
  File "C:\Users\PC\PycharmProjects\trading\venv\lib\site-packages\backtrader\lineiterator.py", line 342, in nextstart
	self.next()
  File "C:/Users/PC/PycharmProjects/trading/scr/test.py", line 17, in next
	bt.TimeFrame.getname(data._timeframe), len(data))
  File "C:\Users\PC\PycharmProjects\trading\venv\lib\site-packages\backtrader\dataseries.py", line 45, in getname
	if compression > 1 or tname == cls.Names[-1]:
TypeError: unorderable types: NoneType() > int()

Ed Bartosh

@nguyễn-tài-nguyên said in Anyone use backtrader to do live trading on Bitcoin exchange?:

@ed-bartosh Hi, when I execute your code, it shows an error. How to fix it ?

You can try to add 'compression=1' to the parameters:

data_min = bt.feeds.CCXT(exchange="bitmex",
                             symbol="BTC/USD",name="btc_usd_min",
                             timeframe=bt.TimeFrame.Minutes,
                             compression=1,
                             fromdate=hist_start_date)

Søren Pallesen

@Ed-Bartosh live feed with Bitmex works (THANKS !!!) however I had to adjust rateLimit (to 15000) and Count (to 100) not get rejected by the API - so maybe you should make these parameters to the bt.feeds.CCXT call.

With Bitmex I can get 1M, 5M, 1H and 1D resolution do you have any ideas how to resample the data for a strategy that needs 2H candles? Is there anyway to manipulate the data feed while running the strategy?

I tried to add the data with cerebro.resampledata but that does not work.

Søren Pallesen

@Ed-Bartosh I would be fantastic if you could fix the historical data load with Bitmex as well so the indicators get their warm up data upfront. Thx again for you help.

Nguyễn Tài Nguyên

@ed-bartosh said in Anyone use backtrader to do live trading on Bitcoin exchange?:

compression=1'

@ed-bartosh said in Anyone use backtrader to do live trading on Bitcoin exchange?:

@nguyễn-tài-nguyên said in Anyone use backtrader to do live trading on Bitcoin exchange?:

@ed-bartosh Hi, when I execute your code, it shows an error. How to fix it ?

You can try to add 'compression=1' to the parameters:
data_min = bt.feeds.CCXT(exchange="bitmex",
                             symbol="BTC/USD",name="btc_usd_min",
                             timeframe=bt.TimeFrame.Minutes,
                             compression=1,
                             fromdate=hist_start_date)

I did it but it still show that error

kazi

Hi @ed-bartosh, I have started testing and I have not been able to get more than a few candles without encountering the following error:

Traceback (most recent call last):
  File "/usr/local/lib/python3.6/site-packages/ccxt/base/exchange.py", line 344, in fetch_requests
    response.raise_for_status()
  File "/usr/local/lib/python3.6/site-packages/requests/models.py", line 935, in raise_for_status
    raise HTTPError(http_error_msg, response=self)
requests.exceptions.HTTPError: 502 Server Error: Bad Gateway for url: https://api.hitbtc.com/api/2/public/candles/EOSBTC?period=M5&limit=450

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "hist-test.py", line 143, in <module>
    sys.exit(runstrategy(sys.argv))
  File "hist-test.py", line 140, in runstrategy
    cerebro.run()
  File "/usr/local/lib/python3.6/site-packages/backtrader/cerebro.py", line 1127, in run
    runstrat = self.runstrategies(iterstrat)
  File "/usr/local/lib/python3.6/site-packages/backtrader/cerebro.py", line 1295, in runstrategies
    self._runnext(runstrats)
  File "/usr/local/lib/python3.6/site-packages/backtrader/cerebro.py", line 1538, in _runnext
    drets.append(d.next(ticks=False))
  File "/usr/local/lib/python3.6/site-packages/backtrader/feed.py", line 404, in next
    ret = self.load()
  File "/usr/local/lib/python3.6/site-packages/backtrader/feed.py", line 476, in load
    _loadret = self._load()
  File "/usr/local/lib/python3.6/site-packages/backtrader/feeds/ccxt.py", line 91, in _load
    self._fetch_ohlcv()
  File "/usr/local/lib/python3.6/site-packages/backtrader/feeds/ccxt.py", line 125, in _fetch_ohlcv
    since=since, limit=limit)[::-1]:
  File "/usr/local/lib/python3.6/site-packages/backtrader/stores/ccxtstore.py", line 87, in retry_method
    return method(self, *args, **kwargs)
  File "/usr/local/lib/python3.6/site-packages/backtrader/stores/ccxtstore.py", line 121, in fetch_ohlcv
    return self.exchange.fetch_ohlcv(symbol, timeframe=timeframe, since=since, limit=limit)
  File "/usr/local/lib/python3.6/site-packages/ccxt/hitbtc2.py", line 671, in fetch_ohlcv
    response = self.publicGetCandlesSymbol(self.extend(request, params))
  File "/usr/local/lib/python3.6/site-packages/ccxt/hitbtc2.py", line 1001, in request
    response = self.fetch2(path, api, method, params, headers, body)
  File "/usr/local/lib/python3.6/site-packages/ccxt/base/exchange.py", line 289, in fetch2
    return self.fetch_requests(request['url'], request['method'], request['headers'], request['body'])
  File "/usr/local/lib/python3.6/site-packages/ccxt/base/exchange.py", line 357, in fetch_requests
    self.handle_rest_errors(e, response.status_code, self.last_http_response, url, method)
  File "/usr/local/lib/python3.6/site-packages/ccxt/base/exchange.py", line 436, in handle_rest_errors
    self.raise_error(error, url, method, exception if exception else http_status_code, response)
  File "/usr/local/lib/python3.6/site-packages/ccxt/base/exchange.py", line 271, in raise_error
    details,
ccxt.base.errors.ExchangeNotAvailable: hitbtc2 GET https://api.hitbtc.com/api/2/public/candles/EOSBTC?period=M5&limit=450 502 Server Error: Bad Gateway for url: https://api.hitbtc.com/api/2/public/candles/EOSBTC?period=M5&limit=450 <html>
<head><title>502 Bad Gateway</title></head>
<body bgcolor="white">
<center><h1>502 Bad Gateway</h1></center>
<hr><center>nginx</center>
</body>
</html>

I have checked whether the site is accessible and found no issue, I am getting roughly 12 ms latency to the api host.

I was wondering how the fromdate= works, is it supposed to get the historical data from eg https://api.hitbtc.com/api/2/public/candles/EOSBTC?period=M5&limit=450 or will it poll for new data? I have noticed that after adding an indicator the script will hang, here is an example:

from datetime import datetime, timedelta
import sys
import backtrader as bt
import backtrader.indicators as btind
import backtrader.feeds as btfeeds

from pandas import bdate_range

class TestStrategy(bt.Strategy):
    
    def __init__(self):
        self.sma = btind.SimpleMovingAverage(self.data, period=10)
        print("initializtion complete")

    def next(self):
        for data in self.datas:
            print('*' * 5, 'NEXT:', bt.num2date(data.datetime[0]), data._name, data.open[0], data.high[0],
                  data.low[0], data.close[0], data.volume[0],
                  bt.TimeFrame.getname(data._timeframe), len(data))
            if not self.getposition(data) and data.close[0] > self.sma[0]:
                order = self.buy(data, exectype=bt.Order.Market, size=10)
            elif self.getposition(data) and data.close[0] < self.sma[0]:
                order = self.sell(data, exectype=bt.Order.Market, size=10)

    def notify_order(self, order):
        print('*' * 5, "NOTIFY ORDER", order)

def runstrategy(argv):
    # Create a cerebro
    cerebro = bt.Cerebro()
    
    # Create data feeds
    hist_start_date = datetime.utcnow() - timedelta(minutes=30)
    data = bt.feeds.CCXT(exchange="hitbtc2", symbol="BTC/USD", name="btc_usd_1m", timeframe=bt.TimeFrame.Minutes, compression=1, fromdate=hist_start_date)
    cerebro.adddata(data)
    print(data[0])
    # Add the strategy
    cerebro.addstrategy(TestStrategy)

    # Run the strategy
    cerebro.run(stdstats=False)

if __name__ == '__main__':
    sys.exit(runstrategy(sys.argv))

This will start printing data after 10mins.

Thank you in advance!