update joblib CI scripts

[pierreglaser]

Fix miscellaneous CI failures.

[codecov]

Codecov Report

Merging #907 into master will decrease coverage by 0.2%.
The diff coverage is n/a.

@@            Coverage Diff             @@
##           master     #907      +/-   ##
==========================================
- Coverage   95.43%   95.23%   -0.21%     
==========================================
  Files          45       45              
  Lines        6459     6459              
==========================================
- Hits         6164     6151      -13     
- Misses        295      308      +13

Impacted Files	Coverage Δ
joblib/disk.py	81.66% <0%> (-6.67%)	⬇️
joblib/backports.py	87.5% <0%> (-6.25%)	⬇️
joblib/_parallel_backends.py	93.6% <0%> (-0.81%)	⬇️
joblib/test/test_hashing.py	98.36% <0%> (-0.55%)	⬇️
joblib/test/test_parallel.py	96.55% <0%> (-0.42%)	⬇️
joblib/test/test_numpy_pickle.py	98.19% <0%> (-0.17%)	⬇️
joblib/logger.py	88.15% <0%> (+1.31%)	⬆️

---

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 b883dc2...f348104. Read the comment docs.

[codecov]

Codecov Report

Merging #907 into master will decrease coverage by 0.07%.
The diff coverage is 100%.

@@            Coverage Diff             @@
##           master     #907      +/-   ##
==========================================
- Coverage   95.43%   95.35%   -0.08%     
==========================================
  Files          45       45              
  Lines        6459     6460       +1     
==========================================
- Hits         6164     6160       -4     
- Misses        295      300       +5

Impacted Files	Coverage Δ
joblib/test/test_dask.py	98.5% <100%> (ø)	⬆️
joblib/disk.py	81.66% <0%> (-6.67%)	⬇️
joblib/_parallel_backends.py	93.6% <0%> (-0.81%)	⬇️
joblib/logger.py	88.15% <0%> (+1.31%)	⬆️

---

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 b883dc2...247d084. Read the comment docs.

[pierreglaser]

The conda solver fails in an undeterministic fashion for the joblib + scikit-learn entry.
Steps to reproduce:

conda update
conda create -n testenv python=3.6 numpy=1.16 pip pytest
conda install scikit-learn

I get an error approx. 25% of the time:

Collecting package metadata (current_repodata.json): done                                                                                                                                                          
Solving environment: failed                                                                                                                                                                                        
Collecting package metadata (repodata.json): done                                                                                                                                                                  
Solving environment: failed                                                                                                                                                                                        
                                                                                                                                                                                                                   
PackagesNotFoundError: The following packages are not available from current channels:                                                                                                                             
                                                                                                                                                                                                                   
  - pkgs/free/linux-64::blas==1.0=mkl                                                                                                                                                                              
                                                                                                                                                                                                                   
Current channels:                                                                                                                                                                                                  
                                                                                                                                                                                                                   
  - https://repo.anaconda.com/pkgs/main/linux-64                                                                                                                                                                   
  - https://repo.anaconda.com/pkgs/main/noarch                                                                                                                                                                     
  - https://repo.anaconda.com/pkgs/free/linux-64                                                                                                                                                                   
  - https://repo.anaconda.com/pkgs/free/noarch                                                                                                                                                                     
  - https://repo.anaconda.com/pkgs/r/linux-64                                                                                                                                                                      
  - https://repo.anaconda.com/pkgs/r/

[pierreglaser]

I proposed an upstream fix to the distributed bit of this PR: dask/distributed#2841

[ogrisel]

Thanks @pierreglaser!