-
-
Notifications
You must be signed in to change notification settings - Fork 1.3k
-
Hi.
I work on BTCUSDT 1 minute interval 3 years data. And using "skopt" optimization max_tries=200. I need optimize 16 parameters.
It takes a lot of time. Around -45Minute for 1 backtesting with this huge data. And gave me really bad results is not worth the time.
Is there anything can we speed up things like using parallel for skopt optimization?
Im using Windows 11. I read the this article: https://kernc.github.io/backtesting.py/doc/examples/Parameter%20Heatmap%20&%20Optimization.html#Model-based-optimization
Beta Was this translation helpful? Give feedback.
All reactions
Replies: 1 comment
-
I suggest installing ubuntu, either as a typical install if you like, or with virtual box and run it through there. it seems that the lib is made to work using fork and spawn is not working properly. and also i think you will be better off only optimizing on 3-6 paramters.
Beta Was this translation helpful? Give feedback.