Can no longer install numba with python 2

[KrisThielemans]

The synergistic demos use brainweb, which starts by using pip install numba. That now fails:

 running bdist_wheel
  /usr/bin/python /tmp/pip-install-sW2ewi/llvmlite/ffi/build.py
    File "/tmp/pip-install-sW2ewi/llvmlite/ffi/build.py", line 122
      raise ValueError(msg.format(_ver_check_skip)) from e
                                                       ^
  SyntaxError: invalid syntax

Also seen here https://stackoverflow.com/questions/61925676/can%c2%b4t-install-numba-for-python with a comment "try a new version of Python".

Is there a work-around? Do we need numba or can we make it optional (without a lot of work).

This is currently a problem as our VM is still stuck on Python 2

[casperdcl]

I think numba is a leftover from some performance hacks @rijobro was trying - looks like it can be removed.

[rijobro]

yeah, it looks like it wasn't actually being used in the notebook. It is, however, used in a few others (but not pip installed):

• https://github.com/SyneRBI/SIRF-Exercises/blob/master/notebooks/Synergistic/de_Pierro_MAPEM.ipynb
• https://github.com/SyneRBI/SIRF-Exercises/blob/master/notebooks/Synergistic/Solutions/Dual_PET_Answer-noMotion.ipynb
• https://github.com/SyneRBI/SIRF-Exercises/blob/master/notebooks/Synergistic/Solutions/Dual_PET_Answer-wMotion.ipynb

[KrisThielemans]

Serves us right from not upgrading to Python 3. @paskino, this will have to be a priority after 2.2 (unless it's a matter of dropping in ubuntu 20.04, but that seems unlikely!)

I suggest to move the import numba to a new cell, but commented out, saying that they can uncomment it and subsequent @jit lines if they have python3 (and numba).

[casperdcl]

How much of a speedup does it really give in those notebooks? May just be better to leave it out...

[rijobro]

Comment from a notebook says

Using jit gets computation time from 90 secs to 2!

So I'm inclined to comment it out rather than delete.