Alejandro de la Vega

+1 on this issue. The weight thing is this didn't used to happen, and now there is a long delay or it requires a password to be typed before it...

@alcrene Did you identify a workaround? Same issue here

I suppose if this is just pulled from the package that it should be fairly harmless. That said, I'm not entirely sure when you'd have access to a class object...

Ah, gotcha. I think when saving out the object it certainly makes sense to annotate it w/ nimads/nimare versions. However, when loading from file (for example, creating a `Dataset` object...

I think if installing locally w/ conda is easy then removing the Docker image is fine. It only makes sense to me for difficult to set up dependencies like mallet.

That's fine w/ me although it seems the biggest maintenance burden is the Dockerfile itself

https://github.com/neurosynth/neurosynth/blob/master/neurosynth/analysis/cluster.py

Numba can optimize numerical computations but is not OOP aware or Pandas aware. Must be pure numpy/basic types to see most gains. It's possible we can extend Numba for sparse...

Yes, I was looking at that, and realized that's the situation. Still feels odd because this transfer is really doing very different things depending on the situation, so it's already...

As an aside, the way the montecarlo methods are written it's also looking tricky to optimize w/ Numba, since the repeated portion is full of Python objects and non-numerical computation....