Gabriel Araujo
Gabriel Araujo
@stfnschfr haah so the neural stuff runs more like an indicator on your strat.. I may just transform `mounirs-ga-version-2` in an indicator and try to combine it with others. PS...
Those shown relativeProfit results are really not backtest results. There are formulas to convert backtest results to a number that is more accurate to evaluate actual strategy success. You can...
Locale files contains data used to construct those plots at `-w` mode. They resumes evolution statistics over time for each locale.
At `Settings.py`, look for `interpreteBacktestProfit` key ;)
@ironexmaiden Hey, size of candles for backtest are aways in minutes. Did you set that option to 'v1'? Should get transparent results... if still has problems send some logs.. @xaviermiss...
Cheers @donkykong017 the failure to load your explitctly selected evaluation dataset was due to a bug. The fix will be released soon, thanks for the report. On the concept I...
Eval dataset is to check if the best settings found are overfitted, only that. At the same time (yeah 20th epoches) the GA evaluates also on evolution dataset (primary), to...
Hey @donkykong017, U think the thing with these 'easy' locales is that their datasets can be scored high with some overfitting, so parameters that lead to a dumb long position...
This path is set at `japonicus/settings/global.toml`.
I never used the docker setup much lol, but from what I recall japonicus should not find gekko.js, it will connect to gekko thru `http://gekko:3000` that gekko's container will expose....