model.Model

model.Model()

A model class to support a variety of different models. This should accept an input matrix or dataframe X and response array y. This class must also implement the following methods:

Notes

The following methods must be implemented:

fit() This method should fit the model and return the fitted model object. The class should also record the fitted model under the self._fitted attribute. That way the user can access the fitted model directly or after the fact. predict(X): This method should generate predicted values from the fitted model on a holdout set and return these predictions as a numpy array.

Methods

Name	Description
fit	Fit the underlying model and return the model object.
predict	Returns the predictions of the fitted model on a hold-out dataset.

fit

model.Model.fit()

Fit the underlying model and return the model object.

This method should also store the model at the self._fitted attribute for direct user access.

Returns

Name	Type	Description
	Any	An arbitrary fitted model.

predict

model.Model.predict(X)

Returns the predictions of the fitted model on a hold-out dataset.

This function should return predictions as a numpy array.

Parameters

Name	Type	Description	Default
X	pd.DataFrame | np.array | Any	Any rectangular data structure that can generate predictions using the np.dot(X, beta) procedure.	required

Returns

Name	Type	Description
	np.array	A one-dimension numpy array with predicted hold-out values.