nolearn.grid_search

grid_search() is a wrapper around sklearn.grid_search.GridSearchCV.

grid_search() adds a printed report to the standard GridSearchCV functionality, so you know about the best score and parameters.

Usage example:

>>> import numpy as np
>>> from nolearn.dataset import Dataset
>>> from sklearn.linear_model import LogisticRegression
>>> data = np.array([[1, 2, 3], [3, 3, 3]] * 20)
>>> target = np.array([0, 1] * 20)
>>> dataset = Dataset(data, target)

>>> model = LogisticRegression()
>>> parameters = dict(C=[1.0, 3.0])
>>> grid_search(dataset, model, parameters)  
parameters:
{'C': [1.0, 3.0]}
...
Best score: 1.0000
Best grid parameters:
    C=1.0,
...