Benchmarks now with py3

[lasofivec]

Changes in benchmarks scripts:

• instead of print(..., file=...) now using the more elegant write(...
• took out the __init__.py file so that the directory is not packages (only useful for devs)
• made default test smaller

[pep8speaks]

Hello @lasofivec! Thanks for updating this PR. We checked the lines you've touched for PEP 8 issues, and found:

There are currently no PEP 8 issues detected in this Pull Request. Cheers! 🍻

Comment last updated at 2019-11-05 17:52:25 UTC

[codecov-io]

Codecov Report

Merging #246 into devel will not change coverage.
The diff coverage is n/a.

@@           Coverage Diff           @@
##            devel     #246   +/-   ##
=======================================
  Coverage   43.93%   43.93%           
=======================================
  Files          70       70           
  Lines       21288    21288           
=======================================
  Hits         9352     9352           
  Misses      11936    11936

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 488b9f3...b4fd4fb. Read the comment docs.

[lasofivec]

Sorry for the horrible commit messages 🙈

[Didou09]

Ok for me, seems cleaner
Are you sure with the f.write(...) ? I agree it is more elegant and pythonic but it seems quite less compact.
I'm ok with both, it's just because you raised the point and we might try to do the same in the future based on this discussion.

Did you try to launch the benchmark again after these changes ?
Since they are not in the unit tests, travis does not check them (unless my memory is failing)

[Didou09]

• took out the __init__.py file so that the directory is not packages (only useful for devs)
• made default test smaller

2 good ideas :-)

[lasofivec]

Ok for me, seems cleaner
Are you sure with the f.write(...) ? I agree it is more elegant and pythonic but it seems quite less compact.
I'm ok with both, it's just because you raised the point and we might try to do the same in the future based on this discussion.

Did you try to launch the benchmark again after these changes ?
Since they are not in the unit tests, travis does not check them (unless my memory is failing)

Yes I tried them, to be on the safe side, could you try them as well ?

[Didou09]

Ok, I coul try them on the IRFM clusters, it works too 👍