new changes to backends, NEURON introduces a lot of problems. about neuronunit HOT 6 CLOSED

I didn't even really change that much. Most of it is your code.

from neuronunit.

These are some of the relevant change logs:

31e623b

31e623b

7520bff

e17a1c2

fe91681

70e3811

31e623b

4aa609c

I also do not believe this is the full extent of relevant changes to models/__init__.py, models/backends.py, tests/__init__.py, however the question of how much change is required to introduce a bug is misleading, as it only takes a single key stroke to introduce a bug.

My proposed solution is that I temporarily revert to an earlier commit, and then merge in these changes when I have time to manage the bug fixes.

I have created a branch stable from 11 days ago (before the merge with scidash/neuronunit), which can run neuronunit/neuronunit/tests/exhaustive_search.py and nsga.py.

I probably should set up a diff between the files models/__init__.py, models/backends.py, tests/__init__.py, in these days old versions of these files and the new changes you have introduced.

from neuronunit.

Feel free to work with your code on russel_dev or your fork, and then we can commit it to scidash/dev when you and I have time to work through it.

from neuronunit.

Yes, I am happy to fix these problems now that I know they are a priority. I should have noted that I can see how the changes you made make the code better. I will work on integration, and finding the minor changes required to get all the functions working together.
.

from neuronunit.

I have adapted my example exhaustive_search.py (however note, this does not mean I have also adapted nsga.py) code to work with the new implementation of the file backends.py, especially with regards to the function def set_run_params(self, **params)

However I stumped by the error below. Inserting the statement: score_type = scores.ZScore, into the InputResistanceTest class definition, does not work, and the class does contain a compute_score method definition.

I am currently trying: print(type(score)) before return score at line 315 in order to check if compute score is somehow returning an typeless score somehow.

Executing test InputResistanceTest on model vanilla a0.03... Traceback (most recent call last):
  File "/opt/conda/lib/python3.5/site-packages/scoop-0.7.2.0-py3.5.egg/scoop/_control.py", line 150, in runFuture
    future.resultValue = future.callable(*future.args, **future.kargs)
  File "/opt/conda/lib/python3.5/runpy.py", line 254, in run_path
    pkg_name=pkg_name, script_name=fname)
  File "/opt/conda/lib/python3.5/runpy.py", line 96, in _run_module_code
    mod_name, mod_spec, pkg_name, script_name)
  File "/opt/conda/lib/python3.5/runpy.py", line 85, in _run_code
    exec(code, run_globals)
  File "exhaustive_search.py", line 92, in <module>
    score_matrixt=list(futures.map(gs.func2map,iter_))#list_of_models,rhstorage))
  File "/opt/conda/lib/python3.5/site-packages/scoop-0.7.2.0-py3.5.egg/scoop/futures.py", line 99, in _mapGenerator
    for future in _waitAll(*futures):
  File "/opt/conda/lib/python3.5/site-packages/scoop-0.7.2.0-py3.5.egg/scoop/futures.py", line 360, in _waitAll
    for f in _waitAny(future):
  File "/opt/conda/lib/python3.5/site-packages/scoop-0.7.2.0-py3.5.egg/scoop/futures.py", line 337, in _waitAny
    raise childFuture.exceptionValue
NotImplementedError: Test InputResistanceTest either implements no compute_score method or provides no score_type with a compute method.

Executing test InputResistanceTest on model vanilla... [2017-06-03 00:15:39,054] launcher  (127.0.0.1:33069) INFO    Root process is done.
[2017-06-03 00:15:39,055] launcher  (127.0.0.1:33069) INFO    Finished cleaning spawned subprocesses.

from neuronunit.

Fixed.

from neuronunit.