[ENH] Validate streamlines pre-LiFE

[dPys]

Maybe we could also add an optional brain mask argument to ensure we're only considering fibers inside the brain?

[codecov]

Codecov Report

Merging #2195 into master will increase coverage by 2.21%.
The diff coverage is 72.72%.

@@            Coverage Diff             @@
##           master    #2195      +/-   ##
==========================================
+ Coverage   89.13%   91.35%   +2.21%     
==========================================
  Files         251      251              
  Lines       32427    32456      +29     
  Branches     3422     3412      -10     
==========================================
+ Hits        28905    29650     +745     
+ Misses       2806     2059     -747     
- Partials      716      747      +31     

Impacted Files	Coverage Δ
dipy/tracking/life.py	94.85% <33.33%> (-3.40%)	⬇️
dipy/tracking/tests/test_life.py	98.11% <87.50%> (-1.89%)	⬇️
dipy/viz/regtools.py	31.49% <0.00%> (ø)
dipy/reconst/tests/test_shm.py	99.41% <0.00%> (+0.12%)	⬆️
dipy/io/utils.py	86.41% <0.00%> (+0.54%)	⬆️
dipy/workflows/stats.py	85.81% <0.00%> (+0.70%)	⬆️
dipy/nn/model.py	92.59% <0.00%> (+1.85%)	⬆️
dipy/nn/tests/test_tf.py	88.46% <0.00%> (+1.92%)	⬆️
dipy/reconst/shm.py	92.59% <0.00%> (+3.33%)	⬆️
... and 11 more

[arokem]

Thanks!

I have a suggestion to make this a little less auto-magical. Also, would you mind adding a test?

[pep8speaks]

Hello @dPys, Thank you for updating !

• In the file dipy/tracking/life.py:

Line 473:13: E128 continuation line under-indented for visual indent

• In the file dipy/tracking/tests/test_life.py:

Line 104:19: E128 continuation line under-indented for visual indent
Line 105:33: E128 continuation line under-indented for visual indent
Line 114:13: E128 continuation line under-indented for visual indent

Comment last updated at 2020-07-23 22:57:02 UTC

[dPys]

Test added

[arokem]

Looks like there's some problem with the fixture you added? https://travis-ci.com/github/dipy/dipy/jobs/362957473#L2288

[dPys]

Hmm that's odd. I'm trying to test this module, but when I run pytest -vvv test_life.py, I get:

Hint: make sure your test modules/packages have valid Python names.
Traceback:
/usr/local/anaconda3/lib/python3.7/site-packages/_pytest/python.py:511: in _importtestmodule
    mod = self.fspath.pyimport(ensuresyspath=importmode)
/usr/local/anaconda3/lib/python3.7/site-packages/py/_path/local.py:701: in pyimport
    __import__(modname)
/usr/local/anaconda3/lib/python3.7/site-packages/_pytest/assertion/rewrite.py:152: in exec_module
    exec(co, module.__dict__)
test_life.py:3: in <module>
    import dipy.core.gradients as grad
../../core/gradients.py:10: in <module>
    from dipy.core.sphere import disperse_charges, HemiSphere
../../core/sphere.py:9: in <module>
    from dipy.reconst.recspeed import remove_similar_vertices
E   ModuleNotFoundError: No module named 'dipy.reconst.recspeed'

Have you encountered this error before? It's strange because no relative imports are used in the file and I can easily import dipy.reconst.recspeed in ipython so the install is good...

[arokem]

Yo @dPys: are you sure you are running this in the same env as the import in python? This looks like what I get when I don't have the cython extensions compiled. Could you please try compiling them again (e.g., python setup.py build_ext --inplace)?

[skoudoro]

This looks like what I get when I don't have the cython extensions compiled

+1, I encounter this when there is an environment problem

Can you rebase your PR? Furthermore, I recommend to keep the test simple so to avoid the use of pytest.parametrize. Can you update your test without this decorator ?
Thank you

[dPys]

Thanks @arokem and @skoudoro , that was totally what it was!

Compiling, rebasing, and will update test per everyone's suggestions... Stay tuned

[dPys]

Done

[arokem]

LGTM. I think we can merge when the CI comes back green.

[skoudoro]

CI's are green, Thank you @dPys! merging