Evolution: allow callback funcs to take in Hamiltonian

[adamcallison]

• convenient to use solved system in callbacks
• if hamiltonian is time-dependent, it will be cached and can be used for free by callbacks

Some repeated code in _setup_callbacks factored out in the process. Existing tests extended to cover these changes.

[jcmgray]

Hey Adam, sorry just very busy with non-work stuff the next few days but this looks nice and I'll try and get to it soon.

Just one brief comment is that python has the inspect module for working out how many arguments a function takes - might simplify things?

[adamcallison]

Hi, Johnnie - no rush, good luck with the next few days :)

I considered going via inspect, but I ultimately decided on try except because I didn't want to cause problems for more general callables that inspect might not like. Are you concerned that my approach might be slowed down the try except block?

I will leave this PR as is for now, happy to pick this discussion back up when things settle down for you :)

[jcmgray]

Hey Adam, so after much wrestling I seem to have brought quimb+tests+travis into harmony again. If you are still interested in this, could you rebase onto/merge the new changes?

[pep8speaks]

Hello @adamcallison! 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! :beers:

Comment last updated at 2020-02-15 15:45:51 UTC

[codecov[bot]]

Codecov Report

Merging #49 into develop will increase coverage by 4.19%. The diff coverage is 97.22%.

@@             Coverage Diff             @@
##           develop      #49      +/-   ##
===========================================
+ Coverage    84.88%   89.08%   +4.19%     
===========================================
  Files           33       33              
  Lines         8615     8627      +12     
===========================================
+ Hits          7313     7685     +372     
+ Misses        1302      942     -360

Impacted Files	Coverage Δ
quimb/evo.py	98.63% <97.22%> (-0.41%)	:arrow_down:
quimb/tensor/tensor_core.py	91.44% <0%> (+0.3%)	:arrow_up:
quimb/linalg/base_linalg.py	90.64% <0%> (+0.58%)	:arrow_up:
quimb/tensor/array_ops.py	82.43% <0%> (+1.35%)	:arrow_up:
quimb/tensor/optimize_autograd.py	87.43% <0%> (+50.26%)	:arrow_up:
quimb/tensor/optimize_tensorflow.py	67.92% <0%> (+67.92%)	:arrow_up:
quimb/tensor/optimize_pytorch.py	73.37% <0%> (+73.37%)	:arrow_up:

---

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 2bcdd4a...f3ef77b. Read the comment docs.

[adamcallison]

Hey Adam, so after much wrestling I seem to have brought quimb+tests+travis into harmony again. If you are still interested in this, could you rebase onto/merge the new changes?

Hi Johnnie, hope all is well. Rebase done (plus an extra commit to make it pep8 compliant)

[jcmgray]

OK, finally got round to reviewing this, all looks good! Thanks for the PEP8 updates and another all round good and thorough pull request.