A simple version of ridge implemented in sklearn or torch, with nested-validation
- Generate toy examples by example.py.
- Inherit from sklearn.base.BaseEstimator and sklearn.base.RegressorMixin to implement a simple ridge regression class called Simple_ridge.
- Wrap the class with sklearn.model_selection.GridSearchCV and sklearn.model_selection.KFold to perform nested-validation.
- Add a parameter to control whether we need to reshape a multi-dimention tensor to a matrix?