integrate ModelicaRes into BuildingsPy

[mwetter]

No description provided.

[mwetter]

Merge pull request #3 to the branch issue4_modelicaRes as some further development is needed before merging it to the master.

[kdavies4]

Let me know if I need to close that pull request on master and reopen on issue4_modelicaRes.

[mwetter]

That would be great; otherwise I can do it next week.

[kdavies4]

It's now pull request #5.

[mwetter]

@kdavies4
I merged the pull request, but the doctests now fail on the branch issue4_modelicaRes. I run

sudo pip install modelicares
mwetter@srg-mw:BuildingsPy$ make doctest
python -m doctest \
    buildingspy/io/*.py \
    buildingspy/examples/*.py \
    buildingspy/examples/dymola/*.py \
    buildingspy/simulate/*.py \
        buildingspy/development/*.py
Traceback (most recent call last):
  File "/usr/lib/python2.7/runpy.py", line 162, in _run_module_as_main
    "__main__", fname, loader, pkg_name)
  File "/usr/lib/python2.7/runpy.py", line 72, in _run_code
    exec code in run_globals
  File "/usr/lib/python2.7/doctest.py", line 2783, in <module>
    sys.exit(_test())
  File "/usr/lib/python2.7/doctest.py", line 2772, in _test
    m = __import__(filename[:-3])
  File "buildingspy/io/outputfile.py", line 4, in <module>
    from modelicares import SimRes
  File "/usr/local/lib/python2.7/dist-packages/modelicares/__init__.py", line 63, in <module>
    from modelicares.linres import LinRes
  File "/usr/local/lib/python2.7/dist-packages/modelicares/linres.py", line 24, in <module>
    from control.matlab import ss
ImportError: No module named control.matlab
make: *** [doctest] Error 255

[kdavies4]

I'm sorry. I meant to respond to this. It should work to do pip install control. In the next version of ModelicaRes (0.11.0), I'll try to get that to install automatically. It seems that I should use setuptools instead of distutils and include an install_requires list).

[mwetter]

Using pip install control works with the commit b14bf0b

There is still one test failing:

FAIL: test_addMethods (test_simulate_Simulator.Test_simulate_Simulator)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/home/mwetter/proj/ldrd/bie/modeling/github/iea-annex60/BuildingsPy/buildingspy/tests/test_simulate_Simulator.py", line 83, in test_addMethods
    np.testing.assert_allclose(0.725, r.mean('source.y'))
  File "/usr/lib/python2.7/dist-packages/numpy/testing/utils.py", line 1168, in assert_allclose
    verbose=verbose, header=header)
  File "/usr/lib/python2.7/dist-packages/numpy/testing/utils.py", line 636, in assert_array_compare
    raise AssertionError(msg)
AssertionError: 
Not equal to tolerance rtol=1e-07, atol=0

(mismatch 100.0%)
 x: array(0.725)
 y: array(0.7250000834465027, dtype=float32)

----------------------------------------------------------------------
Ran 18 tests in 32.801s

FAILED (failures=1)

I want to understand first why there is such large a rounding error before accepting the new result.

[kdavies4]

Is this a test that passed with DyMat before? I'm not sure why there would be rounding error, but I can look into the test case further if necessary.

[mwetter]

Yes, this test passed prior to the ModelicaRes integration. Maybe ModelicaRes computes the integral differently, which it then uses to compute the mean?

[kdavies4]

Yes, I suspect that it could be different. ModelicaRes is using scipy.integrate.trapz whereas I think it was a custom routine before.