Training a Binary Classifier
Training a Binary Classifier · Let’s simplify the problem for now and ü only try to identify one digit —for example, the number 5 . · This “ 5-detector ” will be an example of a binary classifier , ü capable of distinguishing between just two classes, · 5 · not-5. ü Let’s create the target vectors for this classification task : y_train_5 = ( y_train == 5 ) # True for all 5s, False for all other digits y_test_5 = ( y_test == 5 ) · Now let’s pick a classifier and train it . · A good place to s tart is with a ü Stochastic Gradient Descent (SGD) classifier , · using Scikit-Learn’s SGDClassifier class. · This classifier has the advantage of being capable of handling very large datasets efficiently. · This is in part because SGD deals with training instances independently , one at a time. ü Let’s create an SGDClassifier and train it on the whole training set: fr