Using Stochastic Gradient Descent for classification使用随机梯度下降来分类

As was discussed in Chapter 2, Working with Linear Models, Stochastic Gradient Descent is a fundamental technique to fit a model for regression. There are natural connections between the two techniques, as the name so obviously implies.

正如在第二章讨论的运行线性模型，随机梯度下降对于拟合回归模型是一个基本的技术，在两个技术间，像它们的名字指出的那样有些自然联系。

Getting ready准备工作

In regression, we minimized a cost function that penalized for bad choices on a continuous scale, but for classification, we'll minimize a cost function that penalizes for two (or more) cases.

在回归模型中，我们用惩罚在连续范围内的错误选择来最小化代价函数，但是对于分类方法，我们用惩罚两个或多个例子来最小化损失函数。

How to do it…怎么做

First, let's create some very basic data:首先，让我们生成一些非常基本的数据集。

from sklearn import datasets
X, y = datasets.make_classification()

Next, we'll create a SGDClassifier instance:然后，我们将生成一个SGDClassifier例子

from sklearn import linear_model
sgd_clf = linear_model.SGDClassifier()

As usual, we'll fit the model:和以前一样，来拟合模型：

sgd_clf.fit(X, y)
SGDClassifier(alpha=0.0001, average=False, class_weight=None,
              early_stopping=False, epsilon=0.1, eta0=0.0, fit_intercept=True,
              l1_ratio=0.15, learning_rate='optimal', loss='hinge',
              max_iter=1000, n_iter_no_change=5, n_jobs=None, penalty='l2',
              power_t=0.5, random_state=None, shuffle=True, tol=0.001,
              validation_fraction=0.1, verbose=0, warm_start=False)

We can set the class_weight parameter to account for the varying amounts of unbalance in a dataset.

我们能设置class_weight参数来调整数据集中不均衡的变化情况。

The Hinge loss function is defined as follows:以下是定义的hinge损失函数

Here, t is the true classification denoted as +1 for one case and -1 for the other. The vector of coefficients is denoted by y as fit from the model, and x is the value of interest. There is also an intercept for good measure. To put it another way:

这里，t是一个真实的分类标志+1为一类和-1为另一类。系数的矢量被标志模型中拟合的y。x是感兴趣的一个值。一般还有个额外的截距，展示为另一种方法。