Closed Mamba413 closed 1 year ago
Thanks for your practical suggestions!
I have replaced the original parameter seed
by random_state
, hist
by cov_matrix
, k
by s
and lambda_
by alpha
.
I also tested its compatibility with sklearn.model_selection.GridSearchCV
and it runs well for selecting alpha
.
More other compatibility tests will be made later.
See more details in PortfolioSelection.
seed
withrandom_state
? The latter is a convention parameter inscikit-learn
.hist
withcov_matrix
? See this: https://pyportfolioopt.readthedocs.io/en/latest/_modules/pypfopt/efficient_frontier/efficient_frontier.html#EfficientFrontier.__init__k
withs
might be better because we uses
in the other places when illustrating sparsity constraint optimization.lambda_
withlambda
?scikit-learn
? For example, you may test whethersklearn.model_selection.GridSearchCV
can be used for selectinglambda
?