Juan M. Cruz-Martinez

The report has a fixed `Q` all across for plotting (1.65 GeV, which is generally what we use) while the next-iteration-values (vp-nextfitruncard) are taken (correctly now xd) at `Q0`. Now......

Not sure whether we addressed this issue specifically, but we are now running hyperopt normally and not seeing these problems, so I'm going to close this issue for the time...

@enocera @peterkrack please, remember to add the "add new data tutorial"! (not sure whether it is already in some of the PR regarding new data, just want to make sure...

> Question: we're talking about test within nnpdf, right? i.e. on FK tables? No! Yes, within NNPDF but only on the XXX.yaml file. Downloading the theories in the CI and...

To put things in perspective, do `python -c 'import tensorflow'` But ok to all suggestions. Seaborn is only used in the `hyperplot` stuff so it is easy to hide away...

Have it on and off on demand would be really good (so that it could maybe be on for `n3fit` but off for `comparefits`)

I guess that's a different problem to be solved... but I've had to rerun fits because I forgot there was some change in `theroy_53` which makes me sad :__ I...

Ah, yes, sorry. I was always thinking about a flag that looks whether it needs to redownload it!

This is actually the same as #1493

Why not different package versions? Why would `nnpdf=flbasis` not be ok and easier (as automatically ensures that only one nnpdf package is installed) Also, wouldn't also be easier to create...