Class LinearRegression

source code

     object --+        
              |        
  BaseEstimator --+    
                  |    
RegressionEstimator --+
                      |
                     LinearRegression

Linear regression:

    f(beta, X, y) = (1 / (2 * n)) * ||Xbeta - y||²_2,

where ||.||²_2 is the squared L2-norm.

Parameters
----------
algorithm : ExplicitAlgorithm. The algorithm that should be used.
        Should be one of:
            1. GradientDescent(...)

        Default is GradientDescent(...).

algorithm_params : A dict. The dictionary algorithm_params contains
        parameters that should be set in the algorithm. Passing
        algorithm=GradientDescent(**params) is equivalent to passing
        algorithm=GradientDescent() and algorithm_params=params. Default
        is an empty dictionary.

start_vector : BaseStartVector. Generates the start vector that will be
        used.

mean : Boolean. Whether to compute the squared loss or the mean squared
        loss. Default is True, the mean squared loss.

Examples
--------
>>> import numpy as np
>>> import parsimony.estimators as estimators
>>> import parsimony.algorithms.gradient as gradient
>>>
>>> np.random.seed(42)
>>>
>>> n = 10
>>> p = 16
>>> X = np.random.rand(n, p)
>>> y = np.random.rand(n, 1)
>>> lr = estimators.LinearRegression(algorithm=gradient.GradientDescent(),
...                                  algorithm_params=dict(max_iter=1000),
...                                  mean=False)
>>> error = lr.fit(X, y).score(X, y)
>>> print "error = ", error
error =  0.0116466703591