Wouter Kouw

The eScience Center could potentially help with this, as part of both the EviDENce and the inside-the-filter-bubble projects. Note that I can't decide this, I'm just saying that this sounds...

Doesn't this already exist? The `NormalMixture` node? Or do you mean extracting the pdf?

Sorry for the late response; I was on vacation. There are two reasons for inspecting messages: 1) debugging and 2) education. Sepideh showed me an `@call_rule` macro to force RMP...

Possibly. If I understand you correctly, you want to replace the scale factors in `ScaledMessage` with the history object? If that object is exposed to the user, then I can...

Ah, interesting. I suspect that this will work often as a default. But we should include a definition clarification in the documentation.

Hi TissueC, thanks for your interest in the library. I think this is a numerical stability issue. Scipy's kde will only return a divide-by-zero if all estimated weights are 0,...

You have too many samples. TCA is a kernel-based method, meaning its memory cost is quadratic+. The standard approach is to subsample your data; take every 10th, 100th or 1000th...

Complex ValueErrors usually occur when you take log or sqrt of negative numbers. Are you sure your BoW representation doesn't contain negative entries somehow?

Did you use pip? Or conda? The package is registered on pip, but not on conda-forge or other anaconda package repos.

Hi @rchew, yeah sure. I had big plans for this toolbox, but have changed my research focus in the last 2/3 years and I don't think I have time for...