Closed desilinguist closed 2 years ago
Merging #702 (afad8de) into main (4ead823) will increase coverage by
0.03%
. The diff coverage is100.00%
.
@@ Coverage Diff @@
## main #702 +/- ##
==========================================
+ Coverage 96.85% 96.89% +0.03%
==========================================
Files 63 63
Lines 9098 9197 +99
==========================================
+ Hits 8812 8911 +99
Misses 286 286
Impacted Files | Coverage Δ | |
---|---|---|
tests/test_preprocessing.py | 100.00% <ø> (ø) |
|
skll/data/writers.py | 94.17% <100.00%> (+0.06%) |
:arrow_up: |
skll/learner/__init__.py | 97.13% <100.00%> (+0.04%) |
:arrow_up: |
skll/learner/utils.py | 94.62% <100.00%> (+0.03%) |
:arrow_up: |
tests/test_commandline_utils.py | 99.66% <100.00%> (+<0.01%) |
:arrow_up: |
tests/test_featureset.py | 99.78% <100.00%> (+<0.01%) |
:arrow_up: |
tests/test_regression.py | 99.64% <100.00%> (+<0.01%) |
:arrow_up: |
tests/test_cv.py | 100.00% <0.00%> (ø) |
|
tests/test_output.py | 100.00% <0.00%> (ø) |
|
tests/test_classification.py | 100.00% <0.00%> (ø) |
|
... and 14 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 4ead823...afad8de. Read the comment docs.
Should the scikit-learn
version also be updated in the Conda recipe? It is currently >=0.24.1,<=0.24.2
. Would this version include 1.0
, e.g. >=1.0,<=1.0.1
?
Should the
scikit-learn
version also be updated in the Conda recipe? It is currently>=0.24.1,<=0.24.2
. Would this version include1.0
, e.g.>=1.0,<=1.0.1
?
Good catch! I will modify it to be the same as the requirements files.
This PR closes #699.
This change is pretty straightforward but it’s definitely backwards incompatible which we will reflect in the release version when we put together the release.
The change specifically motivated by the upgrade are:
requirements.txt
andconda_requirements.txt
to point to the latest scikit-learn (v1.0.1) and allow up to v1.0.2.squared_error
as the default value of theloss
parameter forRANSACRegressor
.toarray()
for converting sparse numpy arrays to dense instead oftodense()
since the latter returns annp.matrix
instead of annp.ndarray
. Scikit-learn is planning to drop support fornp.matrix
inputs and is already displayingFutureWarning
s.normalize
attribute for linear modelsFalse
by default in most scikit-learn linear models anyway and so no warnings are surfaced in SKLL.Lars
, the default value ofnormalize
is still set toTrue
and so we need to force it to False to avoidFutureWarning
instances.if
statement added will actually lead to an execption in the_create_estimator()
method when thenormalize
attribute doesn't exist, so that will be the perfect reminder to excise it entirely when the time comes.Other changes include: