Wojtek Reise

I wanted to have a look at the notebook, but i do not have access rights- you should receive an email requesting them. For sampling from arbitrary densities, something like...

I think it would greatly help if we kept the list of versions up to date, so as we just aways run update-versions, without updating the list itself in the...

Thank you for reporting this, @malalejandra ! At first sight, it seems that the issue stems from the naming of parameters, and different APIs for giotto and Gudhi (or persim?)....

@ulupo , do you have any thoughts on that? The interface for the wasserstein distance allows to modify all the parameters mentioned here, as desired. Ideally, i would propose to...

I had a pretty thorough look, and imo, it does what it claims to. :) The new additions should be covered by the tests, apart from the plotting method.

> (...) we can't really ship this without intervening in some way on the downstream vectorization methods, even if it's just by saying "this is not supported with extended persistence"...

@ulupo @lewtun What is the desired output in this case? I believe the cover should remain 'open' and I see two possibilities: 1) output no interval (since none of them...

> * Reflect a bit more on whether the current choice of `1e-8` to trigger a warning is correct. Maybe a back-of-the-envelope calculation would show that there is a "provably...

Amazing, this is what was missing! The error regarding `versions` is normal. In fact, we are not using `make html`, but, we have targets: - `make all-gh`- generate the theory,...

Hey @raphaelreinauer , thank you for pointing this out. Let me provide clarification here before the docs are updated. For each dimension, we choose the minimum value which appears in...