Plotting does not work and raises an error with Bokeh

[Umutoni-ga]

Expected Behavior

Using a backtest instance to plot (calling the plot() function) the graphs of the performance of the strategy.

Actual Behavior

I installed the Backtesting framework and ran a simple strategy but when I call plot() on a backtest instance I get a TypeError because of Bokeh

---

TypeError Traceback (most recent call last)
d:\CMU-A\Notebooks\FQAAT\Assignment 3\Bollinger Bands.ipynb Cell 35 line 1
----> 1 backtest.plot()

File c:\Users\Gaella\anaconda3\envs\general\Lib\site-packages\backtesting\backtesting.py:1592, in Backtest.plot(self, results, filename, plot_width, plot_equity, plot_return, plot_pl, plot_volume, plot_drawdown, smooth_equity, relative_equity, superimpose, resample, reverse_indicators, show_legend, open_browser)
1589 raise RuntimeError('First issue backtest.run() to obtain results.')
1590 results = self._results
-> 1592 return plot(
1593 results=results,
1594 df=self._data,
1595 indicators=results._strategy._indicators,
1596 filename=filename,
1597 plot_width=plot_width,
1598 plot_equity=plot_equity,
1599 plot_return=plot_return,
1600 plot_pl=plot_pl,
1601 plot_volume=plot_volume,
1602 plot_drawdown=plot_drawdown,
1603 smooth_equity=smooth_equity,
1604 relative_equity=relative_equity,
1605 superimpose=superimpose,
1606 resample=resample,
1607 reverse_indicators=reverse_indicators,
1608 show_legend=show_legend,
...
348 sizing_mode=sizing_mode,
349 )
351 return gp

TypeError: bokeh.models.tools.Toolbar() got multiple values for keyword argument 'logo'

Steps to Reproduce

1. create a simple strategy class
2. create a backtest object
3. backtest the strategy using run()
4. plot the strategy using plot()

class BBtrategy(Strategy):

    def init(self):
      #some indicator logic
       pass

    def next(self):
      #some trading rules
        pass


backtest = Backtest(eur_usd, BollingerBandsStrategy)
backtest.run()
backtest.plot()

Additional info

• Backtesting version: 0.3.3
• bokeh.__version__: 3.2.2
• OS: Windows-10-10.0.22621-SP0

[scottfrye]

This is a problem with bokeh and they indicate on their github that it has been fixed in Version 3.3 which was due out in Aug 2023

[Quantbrinx]

try pip install bokeh==3.3.0.dev2

[SurpriseMF3000]

Hey, im getting the same error;(
bokeh 3.3.0 didn't work for me.

Are there any solutions?

[AaroncoolPx]

Hey, im getting the same error;( bokeh 3.3.0 didn't work for me.

Are there any solutions?

Same here

[AaroncoolPx]

I realise that the problem that I posted is caused by installed pandas version 2.1. Apparently, the parameter method has been deprecated and removed after pandas 2.0.
By installing pandas 1.5.2, it is able to plot.
pip install pandas==1.5.2

[AaroncoolPx]

Not sure how to fix the code, any ideas?

File ~/.local/lib/python3.10/site-packages/backtesting/_plotting.py:148, in _maybe_resample_data.<locals>._group_trades.<locals>.f(s, new_index, bars)
    147 mean_time = int(bars.loc[s.index].view(int).mean())
--> 148 new_bar_idx = new_index.get_loc(mean_time, method='nearest')
    149 return new_bar_idx

TypeError: Index.get_loc() got an unexpected keyword argument 'method'

[DominikBerger01]

Not sure how to fix the code, any ideas?

File ~/.local/lib/python3.10/site-packages/backtesting/_plotting.py:148, in _maybe_resample_data.<locals>._group_trades.<locals>.f(s, new_index, bars)
    147 mean_time = int(bars.loc[s.index].view(int).mean())
--> 148 new_bar_idx = new_index.get_loc(mean_time, method='nearest')
    149 return new_bar_idx

TypeError: Index.get_loc() got an unexpected keyword argument 'method'

Your error differs from the one above. Refer to this issue where yours has been solved: #987

[DominikBerger01]

Expected Behavior

Using a backtest instance to plot (calling the plot() function) the graphs of the performance of the strategy.

Actual Behavior

I installed the Backtesting framework and ran a simple strategy but when I call plot() on a backtest instance I get a TypeError because of Bokeh

TypeError Traceback (most recent call last) d:\CMU-A\Notebooks\FQAAT\Assignment 3\Bollinger Bands.ipynb Cell 35 line 1 ----> 1 backtest.plot()

File c:\Users\Gaella\anaconda3\envs\general\Lib\site-packages\backtesting\backtesting.py:1592, in Backtest.plot(self, results, filename, plot_width, plot_equity, plot_return, plot_pl, plot_volume, plot_drawdown, smooth_equity, relative_equity, superimpose, resample, reverse_indicators, show_legend, open_browser) 1589 raise RuntimeError('First issue backtest.run() to obtain results.') 1590 results = self._results -> 1592 return plot( 1593 results=results, 1594 df=self._data, 1595 indicators=results._strategy._indicators, 1596 filename=filename, 1597 plot_width=plot_width, 1598 plot_equity=plot_equity, 1599 plot_return=plot_return, 1600 plot_pl=plot_pl, 1601 plot_volume=plot_volume, 1602 plot_drawdown=plot_drawdown, 1603 smooth_equity=smooth_equity, 1604 relative_equity=relative_equity, 1605 superimpose=superimpose, 1606 resample=resample, 1607 reverse_indicators=reverse_indicators, 1608 show_legend=show_legend, ... 348 sizing_mode=sizing_mode, 349 ) 351 return gp

TypeError: bokeh.models.tools.Toolbar() got multiple values for keyword argument 'logo'

Steps to Reproduce

1. create a simple strategy class

2. create a backtest object

3. backtest the strategy using run()

4. plot the strategy using plot()

class BBtrategy(Strategy):

    def init(self):
      #some indicator logic
       pass

    def next(self):
      #some trading rules
        pass


backtest = Backtest(eur_usd, BollingerBandsStrategy)
backtest.run()
backtest.plot()

Additional info

* Backtesting version: 0.3.3

* `bokeh.__version__`: 3.2.2

* OS: Windows-10-10.0.22621-SP0

I was able to reproduce your error. Backtesting version: 0.3.3 doesn't seem to be compatible with latest bokeh versions.
To fix the issue I have downgraded to bokeh==2.4.3 and was able to run your sample code.

Step 1 (Given you are using pip as your package manager)
pip uninstall bokeh

Step 2
pip install bokeh==2.4.3

Now you should be able to run your code just fine. Note other versions of bokeh might work fine as well but 2.4.3 is the one I've been using ever since without issues.

Hope that helps.

Regards,
Dominik

[Umutoni-ga]

@DominikBerger01 It works, thank you very much!

[kernc]

Bokeh issues fixed (e.g. in dfba461). Fixes released.