I made a post in June - Re: cerebro.plot / matplot quality. It was low priority so I finally found a solution and though I would share it with others. Place the following code in the notebooks that you would like higher quality plots:

%config InlineBackend.figure_format = 'svg'

here are all the options ('png') is the default:
‘png’, ‘retina’, ‘jpeg’, ‘svg’, ‘pdf’

I have been consider purchasing data and moving my current strategy testing away from QuantConnect to something local for greater speed and efficient use of computing resources. I have backtrader up and running, yet options seem like they would require a major work around, unless I am missing something.

Thanks!

I made a post in June - Re: cerebro.plot / matplot quality. It was low priority so I finally found a solution and though I would share it with others. Place the following code in the notebooks that you would like higher quality plots:

%config InlineBackend.figure_format = 'svg'

here are all the options ('png') is the default:
‘png’, ‘retina’, ‘jpeg’, ‘svg’, ‘pdf’

I tied both on two different mac machines, both running jupyter notebooks or just running the script.
Thanks!

How do I get the graphic from matplot lib to look as nice as printed on Visual Inspection: Plotting?. I even used the exact script in the docs and my plot looks like a bad 8 bit imitation.

I tried seaborn, not sure what is missing.

@backtrader Got it, I will check it out.

Sorry for the general nature of my question. I have not built anything that complex in backtrader while in both Qantopian and QuantConnect I have, with some help. In QuantConnect I am able to use the built in function OptionChainProvider and AddOptionContract, then us a custom function filter the contracts. I just don't see anything prebuilt in backtrader so I did not know if there was a "easy" system / way that the community was using it to test and trade options.

I have been bumping into the python speed issues but did not realize it had to do with the language itself. Interesting to learn.

Thanks!

I have been consider purchasing data and moving my current strategy testing away from QuantConnect to something local for greater speed and efficient use of computing resources. I have backtrader up and running, yet options seem like they would require a major work around, unless I am missing something.

Thanks!