CODEWITHSUNDEEP
pythonscikit-learnjupyter-notebook
afettouhi
afettouhi

Reputation: 85

mean_validation_score gives an AttributeError

I am currently doing some exercises with Kernel Density Estimation and I am trying to run this piece of code:

from sklearn.datasets import load_digits
from sklearn.model_selection import GridSearchCV

digits = load_digits()

bandwidths = 10 ** np.linspace(0, 2, 100)
grid = GridSearchCV(KDEClassifier(), {'bandwidth': bandwidths}, cv=3)
grid.fit(digits.data, digits.target)

scores = [val.mean_validation_score for val in grid.cv_results_]

but as the title says I get an

---------------------------------------------------------------------------
AttributeError                            Traceback (most recent call last)
<ipython-input-29-15a5f685e6d6> in <module>
      8 grid.fit(digits.data, digits.target)
      9 
---> 10 scores = [val.mean_validation_score for val in grid.cv_results_] 

<ipython-input-29-15a5f685e6d6> in <listcomp>(.0)
      8 grid.fit(digits.data, digits.target)
      9 
---> 10 scores = [val.mean_validation_score for val in grid.cv_results_] 
AttributeError: 'str' object has no attribute 'mean_validation_score'

regarding mean_validation_score and I don't understand why. The code is directly out of a book with a few changes due running an up to date scikit learn package. Here is the original code snipet:

from sklearn.datasets import load_digits
from sklearn.grid_search import GridSearchCV

digits = load_digits()

bandwidths = 10 ** np.linspace(0, 2, 100)
grid = GridSearchCV(KDEClassifier(), {'bandwidth': bandwidths})
grid.fit(digits.data, digits.target)

scores = [val.mean_validation_score for val in grid.grid_scores_]

EDIT:

Forgot to add how bandwiths is defined:

from sklearn.base import BaseEstimator, ClassifierMixin


class KDEClassifier(BaseEstimator, ClassifierMixin):
    """Bayesian generative classification based on KDE

    Parameters
    ----------
    bandwidth : float
        the kernel bandwidth within each class
    kernel : str
        the kernel name, passed to KernelDensity
    """
    def __init__(self, bandwidth=1.0, kernel='gaussian'):
        self.bandwidth = bandwidth
        self.kernel = kernel

    def fit(self, X, y):
        self.classes_ = np.sort(np.unique(y))
        training_sets = [X[y == yi] for yi in self.classes_]
        self.models_ = [KernelDensity(bandwidth=self.bandwidth,
                                        kernel=self.kernel).fit(Xi)
                        for Xi in training_sets]
        self.logpriors_ = [np.log(Xi.shape[0] / X.shape[0])
                            for Xi in training_sets]
        return self

    def predict_proba(self, X):
        logprobs = np.array([model.score_samples(X)
                                for model in self.models_]).T
        result = np.exp(logprobs + self.logpriors_)
        return result / result.sum(1, keepdims=True)

    def predict(self, X):
        return self.classes_[np.argmax(self.predict_proba(X), 1)]

Upvotes: 1

Views: 456

Answers (2)

Palash Mondal
Palash Mondal

Reputation: 538

It's simple, I also face the same problem, Just replace this line-

scores = [val.mean_test_score for val in grid.cv_results_]

with

scores = grid.cv_results_.get('mean_test_score').tolist()

Because, 'mean_test_score' is depricated and grid.cv_results_ is in dict format.

Upvotes: 1

Angelo
Angelo

Reputation: 655

The documentation of the object GridSearchCV specifies that the attribute cv_results_ is a dictionary, therefore, iterating over a python dictionary returns the strings of the keys as you can se here.

My recommendation is to specify at the GridSearchCV constructor the scoring you want to use and then have a look at the cv_results_ dictionary.

Hope it helps.

Upvotes: 0

Related Questions