<br />
<b>Warning</b>:  Declaration of action_plugin_tablewidth::register(&$controller) should be compatible with DokuWiki_Action_Plugin::register(Doku_Event_Handler $controller) in <b>/s/bach/b/class/cs545/public_html/fall15/lib/plugins/tablewidth/action.php</b> on line <b>93</b><br />
===== Multi-class classification in scikit-learn =====

Let's use a One-vs-the-rest classifier on the [[https://archive.ics.uci.edu/ml/datasets/Iris | iris dataset]].  The data has four features that describe features of three types of iris flowers.

<code python>
In [1]: import numpy as np

In [2]: from sklearn import datasets

In [3]: from sklearn.multiclass import OneVsRestClassifier,OneVsOneClassifier

In [4]: from sklearn.svm import LinearSVC,SVC

In [5]: from sklearn import cross_validation

In [6]: iris = datasets.load_iris()

In [7]: X, y = iris.data, iris.target

In [8]: classifier = OneVsRestClassifier(LinearSVC())

In [9]: print np.mean(cross_validation.cross_val_score(classifier, X, y, cv=5))
0.966666666667

In [10]: classifier = OneVsOneClassifier(LinearSVC())

In [11]: print np.mean(cross_validation.cross_val_score(classifier, X, y, cv=5))0.98

In [12]: # does this mean that oneVsOne is better?  not necessarily...

In [13]: classifier = OneVsRestClassifier(SVC(C=1, kernel='rbf', gamma=0.5))

In [14]: print np.mean(cross_validation.cross_val_score(classifier, X, y, cv=5))0.98

</code>

And here's the code without the python prompts to get in the way:

<file python multi_class.py>

import numpy as np
from sklearn import datasets
from sklearn.multiclass import OneVsRestClassifier,OneVsOneClassifier
from sklearn.svm import LinearSVC,SVC
from sklearn import cross_validation
iris = datasets.load_iris()
X, y = iris.data, iris.target

classifier = OneVsRestClassifier(LinearSVC())

print np.mean(cross_validation.cross_val_score(classifier, X, y, cv=5))

classifier = OneVsOneClassifier(LinearSVC())

print np.mean(cross_validation.cross_val_score(classifier, X, y, cv=5))

# does this mean that oneVsOne is better?  not necessarily...

classifier = OneVsRestClassifier(SVC(C=1, kernel='rbf', gamma=0.5))

print np.mean(cross_validation.cross_val_score(classifier, X, y, cv=5))


</file>