Lukas Hergt

Huh, that's curious. Turns out that these test cases depend a lot on the choice of scale/std... I've reduced the example to two parameters. # Case 1: this works fine...

Nice test! And indeed, converting the nested samples from Case 2 to MCMC samples and then doing the importance sampling gets you to the correct posterior: ```python mc7 = MCMCSamples(data=ns7[['a',...

> @AdamOrmondroyd Cant remember what needs doing here? is it just the tests need tidying? and master merging? This also needed waiting for the margarine docs to work, which is...

In light of the rather fast incrementing version number, I'd suggest giving it some more time... Maybe ~o(3months) from 2.0.0 release?

> @lukashergt @AdamOrmondroyd @DilyOng: at the moment we have a .stats function for NestedSamples. The tension_stats function assumes NestedSamples implicitly by calling this, but in practice one can compute logL_p,...

Thanks @brinckmann and @dchooper, > However, the Planck 2018 clik version is _much_ faster. The gain in execution time is incredible, I really recommend switching to planck 2018 (which is...

Switching between editor and pdf viewer is now possible with a mouse click. However, it would be nice if there were a keyboard shortcut/hotkey for this, too.

Thanks for the quick reply, Philipp! Haha, sorry should have directly tried to strip it down, then my question would have been more specific... So, as it turns out for...

Both of your suggestions seem to work. Thanks, Philipp, that way I can keep my blank lines.

> sections in the `text_sections` attribute assume that they end either by the next section or the end of the docstring Good to know, I'll look out for that, but...