Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

'HybridRouteLayer' object has no attribute 'fit' and evaluate #232

Open
shankara-93 opened this issue Apr 4, 2024 · 3 comments
Open

'HybridRouteLayer' object has no attribute 'fit' and evaluate #232

shankara-93 opened this issue Apr 4, 2024 · 3 comments
Labels
enhancement Enhancement to existing features

Comments

@shankara-93
Copy link

HybridRouteLayer lacks built-in functionalities such as fit or evaluate. How can this limitation be addressed?

@andreped
Copy link
Contributor

andreped commented Apr 5, 2024

Hello, @shankara-93! :]

The fit() method in RouteLayer only attemps to find the optimal thresholds. I have yet to see that it improves performance for my applications, so I would think the SR team has not put that much effort into adding it for the HybridRouteLayer class, which is more recent.

I guess this feature could be added to HybridRouteLayer as well, but then I would try to use inheritance here between the two classes. I would think this method will be very similar, if not identical, for the HybridRouteLayer.

@jamescalam, @bruvduroiu: I can draft a PR now if this feature is of interest :]

@jamescalam
Copy link
Member

@andreped I think it'd be great to add fit() to HybridRouteLayer — at some point, we'll create a BaseRouteLayer and base both from this, but we are happy with a more direct solution in the meantime. Let us know if/how we can help!

@jamescalam jamescalam added the enhancement Enhancement to existing features label Apr 7, 2024
@andreped
Copy link
Contributor

andreped commented Apr 7, 2024

@jamescalam Sure. I can make an attempt later today :]

I was thinking of introducing class weights to the pooling method as well, which would enable developers to introduce a priori biases to certain classes. We have one application where this would be of interest already.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
enhancement Enhancement to existing features
Projects
None yet
Development

No branches or pull requests

3 participants