Bart van Erp
Bart van Erp
The `UniformScaling` object is nice to work with but causes issues in de BFE computation. In the average energy computations in some cases it becomes difficult to compute certain terms....
When directly plugging in numbers or fixed values into the model specification, this will send out messages as expected, but these messages do not include scale factors. For example: ```julia...
In the future it would be good to have some kind of benchmarking system in our CI, such that we become aware how changes in our code impact performance. An...
Not sure whether it should be mentioned here or in `ReactiveMP.jl`, but for certain applications it is highly beneficial to support inplace operations (rules, computations, basically everything). In my experiments...
Ran into an issue that I think a lot of people will get who want to try out the `rxinference()` function. For the simple model ```julia @model function observation_model(Hs, Qs,...
It would be a very nice feature to be able to update parameters/datavars/constants in-between iterations, such that we can do annealing and such.
The `PointMass` constraint without a mean-field factorization breaks the BFE computation. I think this issue is related to #32. ### Minimal working examples: #### Works fine ```julia @model function test_model()...
In `RxInfer`, constraints are enforced after the product of two messages as `f(q(x))`. For sum-product message passing, this means that first all messages in the graph are computed, after which...