Albert

Results 34 comments of Albert

what's the progress status of this PR?

Agree with @bartvanerp that `cholinv` should be applied to Hermitian matrices only. However, we can easily encounter situations when the pos-def property is violated. (u) | (x_t_prev) --> + -->...

This isn't the first time we suffer from optimization packages. Should we use our handcrafted optimization procedures or we fix the version of optimization packages? Your thoughts guys, @ismailsenoz @bvdmitri...

Good job @abpolym! Once @bvdmitri comments are addressed, we can merge!

@ismailsenoz haven't you ported the node actually?

Hi @nathanielvirgo! Thanks for trying out `ReactiveMP`. In short, the answer to your question is _no_. At the moment `ReactiveMP` works mainly with conjugate models. In the first example, when...

Adding these rules is definitely a good idea. I would maybe refrain from using the `EqMultiply` node for now. I am not sure that the problem you've mentioned is easy...

I didn't get how you wanted to proceed with it. Shouldn't you PR to `delta` branches, after which we adopt rules and tests?

Nice! The specification of the graph is more relaxed. Although, I don't like the proposed specificator syntax.