This is a fairly old topic but I wondered if there had been any updates on if this has changed?

The advise functions seem fairly limited as they are. The manual doesn't describe them particularly well and it seems like they are only able to enter without control or exit?

Is there any documentation that shows more details on how the advise functions work more than the manual page?

Specifically what they return? If you take the DeepLearn script as an example they open whenever either long or shrot is above the 0.5 threshold, which suggests they both return positive when the theoretical prediction is either 0 or 1?

For example projects I've worked on in the past would classify the prediction into 4 categories, buy, sell, close buy and close sell. Maybe by givin outputs of 1, 0.5, -0.1 and -1.

Sorry if I've missed something in the manual.

Thanks

Tom