You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I am an enthusiast user of this library, but I found an interesting generalization of LRP-alpha-beta (Relative Attributing Propagation) that would be interesting to be supported by iNNvestigate.
The implementation is in pytorch, but maybe it would not be too difficult to port it to the iNNvestigate framework. It would be very useful and it will make the library richer.
I would try to do it myself but I am still not too practical with the framework...
What do you think about it?
PS: Thanks again for this amazing library!!
The text was updated successfully, but these errors were encountered:
Thanks for the suggestion and sorry for the late reply.
I think it is a great idea and as I can see you already implemented it.
Hopefully albermax will merge it soon.
Ah I see, the pull request is not from you, sorry.
Yeah, we will have a look at the implementation. But right now, we are still working on the basic structure of Innvestigate 2.0 for tf 2.0
Hi everybody,
I am an enthusiast user of this library, but I found an interesting generalization of LRP-alpha-beta (Relative Attributing Propagation) that would be interesting to be supported by iNNvestigate.
https://github.com/wjNam/Relative_Attributing_Propagation
The implementation is in pytorch, but maybe it would not be too difficult to port it to the iNNvestigate framework. It would be very useful and it will make the library richer.
I would try to do it myself but I am still not too practical with the framework...
What do you think about it?
PS: Thanks again for this amazing library!!
The text was updated successfully, but these errors were encountered: