In a near future you can. As you said it, you can use
log_long, however to initiate the stuff, I'll add a new method to the API soon called for example
init() which would be the place to initiate third party libraries and then you could use
end() method for saving progress at the end of the backtest. I'm not sure if I've added
end() to the docs yet.
I will also add more output options like
csv for trades, so that you can run backtest once, and then use that output for training your models.