CODEWITHSUNDEEP
pythonmachine-learningscikit-learnclassificationsvm
Christel Junco
Christel Junco

Reputation: 133

SVM Algorithm: Without using sklearn package (Coded From the Scratch)

I'm trying to code SVM algorithm from the scratch without using sklearn package, now I want to test the accuracy score of my X_test and Y_predict. The sklearn had already function for this:

clf.score(X_test,Y_predict)

Now, I traced the code from the sklearn package, I cannot find how the 'score' function has coded from the scratch.

And how the model generated from the sklearn SVC:

SVM classifier :: SVC(C=1.0, cache_size=200, class_weight=None, coef0=0.0, decision_function_shape='ovr', degree=3, gamma=2, kernel='poly', max_iter=-1, probability=False, random_state=None, shrinking=True, tol=0.001, verbose=False)

After I fitted and trained the dataset, I want that the model will be generated so that I can Save and Load it using Pickle.

Upvotes: 1

Views: 3168

Answers (1)

Jacques Kvam
Jacques Kvam

Reputation: 3076

If you use IPython you can usually find out where functions are defined with by appending ?? to the function. For example:

>>> from sklearn.svm import SVC
>>> svc = SVC()
>>> svc.score??
Signature: svc.score(X, y, sample_weight=None)
Source:   
    def score(self, X, y, sample_weight=None):
        """Returns the mean accuracy on the given test data and labels.

        In multi-label classification, this is the subset accuracy
        which is a harsh metric since you require for each sample that
        each label set be correctly predicted.

        Parameters
        ----------
        X : array-like, shape = (n_samples, n_features)
            Test samples.

        y : array-like, shape = (n_samples) or (n_samples, n_outputs)
            True labels for X.

        sample_weight : array-like, shape = [n_samples], optional
            Sample weights.

        Returns
        -------
        score : float
            Mean accuracy of self.predict(X) wrt. y.

        """
        from .metrics import accuracy_score
        return accuracy_score(y, self.predict(X), sample_weight=sample_weight)
File:      ~/miniconda/lib/python3.6/site-packages/sklearn/base.py
Type:      method

In this case it's coming from the ClassifierMixin so this code can be used with all classifiers.

https://github.com/scikit-learn/scikit-learn/blob/master/sklearn/base.py#L310

https://ipython.readthedocs.io/en/stable/interactive/python-ipython-diff.html#accessing-help

Upvotes: 1

Related Questions