Investigate time taken by our implementation

[QB3]

Next steps:

• [ ] always work with w_dense and ws
• [ ] make X[ws] @ w[ws] faster

[codecov-commenter]

Codecov Report

Merging #61 (bf4b02c) into master (b8ce7d3) will increase coverage by 7.40%. The diff coverage is 65.00%.

@@            Coverage Diff             @@
##           master      #61      +/-   ##
==========================================
+ Coverage   56.19%   63.60%   +7.40%     
==========================================
  Files          12       12              
  Lines        1098      783     -315     
  Branches      242      121     -121     
==========================================
- Hits          617      498     -119     
+ Misses        406      254     -152     
+ Partials       75       31      -44     

Impacted Files	Coverage Δ
andersoncd/tests/test_docstring_parameters.py	75.36% <ø> (+0.72%)	:arrow_up:
andersoncd/toprof.py	23.80% <23.80%> (ø)
andersoncd/penalties.py	44.76% <44.76%> (ø)
andersoncd/datafits.py	50.70% <50.70%> (ø)
andersoncd/solver.py	64.02% <64.02%> (ø)
andersoncd/data/synthetic.py	72.41% <69.23%> (-18.50%)	:arrow_down:
andersoncd/estimators.py	92.47% <92.47%> (ø)
andersoncd/__init__.py	100.00% <100.00%> (ø)
andersoncd/data/__init__.py	100.00% <100.00%> (ø)
andersoncd/tests/test_estimators.py	100.00% <100.00%> (ø)
... and 2 more

---

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 ad549a3...bf4b02c. Read the comment docs.

[QB3]

Tests pass including the weighted Lasso. Performance were improved and are almost as good as celer at the first iterations for large values of lambda, and are better for a large number of iterations and small values of lambda.

[mathurinm]

perfect !!!