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
Sanity tests for BitCircuitPair shows some failures (but getting correct results for expectations, need to investigate more)
Tests for helper functions
Leaf cleanup (see comment 1)
Separate out pr-constrain (fvalues) and expectation (fgvalues) part maybe? If it does not slow down too much. In any case, having pr-constraint by itself is useful to have.
Easy and Efficient Support for batching (by reusing memory etc)
See commnet 5, seems much faster
Refactoring
make it easy to use other kernels than expecatations
Moments
Pruning some pairs, some pairs might always be zeros, if we can gurantee that maybe we can remove some. Might help in reducing size.
Specialized Cases: For example if both cases circuits are identical.
Generalized Product
The text was updated successfully, but these errors were encountered:
Code used for demo benchmarking:
https://gist.github.com/khosravipasha/2e56013040be579993fbb65ed56f853d
Remaining tasks, check Tractables/ProbabilisticCircuits.jl#33 for more details.
Sanity tests for BitCircuitPair shows some failures (but getting correct results for expectations, need to investigate more)
Tests for helper functions
Leaf cleanup (see comment 1)
Separate out pr-constrain (fvalues) and expectation (fgvalues) part maybe? If it does not slow down too much. In any case, having pr-constraint by itself is useful to have.
Easy and Efficient Support for batching (by reusing memory etc)
Refactoring
Moments
Pruning some pairs, some pairs might always be zeros, if we can gurantee that maybe we can remove some. Might help in reducing size.
Specialized Cases: For example if both cases circuits are identical.
Generalized Product
The text was updated successfully, but these errors were encountered: