mathurinm commented 3 years ago

Seems fixed, but we have a super slow test:

    model_name = 'svr'
    algo = list_algos[1]
    criterion = 'MSE'
    test_check_grad_sparse_ho(model_name, criterion, algo)

any idea what's going on @Klopfe ?

QB3 commented 3 years ago

Just to keep track of what is happening: the problem is that some tests introduced in the SVR are slow. The goal of the PR is to understand why and to fix this problem.

mathurinm commented 3 years ago

I merged the fixes for #101, the test is still slow

codecov-io commented 3 years ago

Codecov Report

Merging #99 (9ce97a2) into master (97b4299) will increase coverage by 0.00%. The diff coverage is 100.00%.

@@           Coverage Diff           @@
##           master      #99   +/-   ##
=======================================
  Coverage   79.02%   79.03%           
=======================================
  Files          41       41           
  Lines        3037     3038    +1     
  Branches      326      326           
=======================================
+ Hits         2400     2401    +1     
  Misses        544      544           
  Partials       93       93

Flag	Coverage Δ
unittests	`∅ <ø> (∅)`

Flags with carried forward coverage won't be shown. Click here to find out more.

Impacted Files	Coverage Δ
sparse_ho/algo/implicit.py	`91.66% <ø> (ø)`
sparse_ho/tests/test_models.py	`81.18% <ø> (ø)`
sparse_ho/models/svr.py	`80.16% <100.00%> (ø)`
sparse_ho/tests/common.py	`100.00% <100.00%> (ø)`

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 97b4299...9ce97a2. Read the comment docs.

mathurinm commented 2 years ago

closing as we have started to address the issue in another sequence of PRs

QB3 / sparse-ho

[FIX] Fix svr test #99

Codecov Report