According to the Geron book, for multi-class classification,
SGDClassifier in scikit-learn uses one-vs-rest. But how can I tell which one is used as it doesn't appear to give this information in the help file.
It's there in the User guide [Link]
"....SGDClassifier supports multi-class classification by combining multiple binary classifiers in a “one versus all” (OVA) scheme. For each of the classes, a binary classifier is learned that discriminates between that and all other classes. At testing time, we compute the confidence score (i.e. the signed distances to the hyperplane) for each classifier and choose the class with the highest confidence...."