The premise is that mine is just an educated guess; as you can see here there's an ongoing attempt to enrich the documentation related to the use of n_jobs
.
Nevertheless, the answer might be found in what is written in the docs for cross-validation estimators:
Some example of cross-validation estimators are ElasticNetCV and LogisticRegressionCV. Cross-validation estimators are named EstimatorCV and tend to be roughly equivalent to GridSearchCV(Estimator(), ...). The advantage of using a cross-validation estimator over the canonical estimator class along with grid search is that they can take advantage of warm-starting by reusing precomputed results in the previous steps of the cross-validation process. This generally leads to speed improvements. An exception is the RidgeCV class, which can instead perform efficient Leave-One-Out CV.
Basically, the use of RidgeCV
slightly differs from the use of the other cross-validation estimators (among which, for instance, LogisticRegressionCV
, LassoCV
, ElasticNetCV
).
- the former (whenever used with default
cv=None
) implements Ridge regression with built-in Leave-one-out Cross-Validation; whenever cv is not None
, instead, it implements GridSearchCV(Ridge())
with default n_jobs=None
.
- the latter ones do implement more standard cv-strategies with the advantages described above with respect to the use of
GridSearchCV(Estimator())
.
Eventually, some other useful information might be found in this recent thread.
与恶龙缠斗过久,自身亦成为恶龙;凝视深渊过久,深渊将回以凝视…