使用sklearn中svm做多分类时难点解惑

一，parameters: decision_function_shape:
两种方法one v one 或者 one v rest

decision_function_shape : ‘ovo’, ‘ovr’ or None, default=None Whether
to return a one-vs-rest (‘ovr’) decision function of shape (n_samples,
n_classes) as all other classifiers, or the original one-vs-one
(‘ovo’) decision function of libsvm which has shape (n_samples,
n_classes * (n_classes - 1) / 2). The default of None will currently
behave as ‘ovo’ for backward compatibility and raise a deprecation
warning, but will change ‘ovr’ in 0.19. New in version 0.17:
decision_function_shape=’ovr’ is recommended. Changed in version 0.17:
Deprecated decision_function_shape=’ovo’ and None.

二、Attributes:

support_: 返回training data中的support vector的索引（几乎没什么用，因为索引是按大小顺序排列，不知道其属于哪个类别的support vector）

support_vectors_ :返回support_vector的值，而且按照每个类别的support_vector依次排列。

n_support_：每个类别的support_vector个数，对照support_vector使用
如如果返回[3,4,5]，则表明support_vector前三个元素是第一类的sv,4到7个元素是第二类的sv，最后5个是第三类的sv。

dual_coef_ : kernel前面的系数部分。即下图中的an*tn

intercept_: 下图中的b

三，method
decision_function: 返回每个data到每个classifier的距离，注意出现overlap时，support_vector 到其对应classifier 的距离不一定是1

详情请参看bishop 《PRML》333-334页

另外注意：在这个模型中的support vector均对应与class，而不是对应于classifier
即如果data属于class时，它一定是它的boundray或者outliner。但属于classifer时，不清楚它属于classfier中两个class中的哪一个

为了更好理解模型，可以参看stackoverflow大神写的代码：
http://stackoverflow.com/questions/20113206/scikit-learn-svc-decision-function-and-predict

# I've only implemented the linear and rbf kernels#sv:support vector  nv:上面的n_support_  a:上面的dual_coef#b:上面的intercept_ def kernel(params, sv, X):    if params.kernel == 'linear':        return [np.dot(vi, X) for vi in sv]    elif params.kernel == 'rbf':        return [math.exp(-params.gamma * np.dot(vi - X, vi - X)) for vi in sv]# This replicates clf.decision_function(X)def decision_function(params, sv, nv, a, b, X):    # calculate the kernels    k = kernel(params, sv, X)    # define the start and end index for support vectors for each class    start = [sum(nv[:i]) for i in range(len(nv))]    end = [start[i] + nv[i] for i in range(len(nv))]    # calculate: sum(a_p * k(x_p, x)) between every 2 classes    c = [ sum(a[ i ][p] * k[p] for p in range(start[j], end[j])) +          sum(a[j-1][p] * k[p] for p in range(start[i], end[i]))                for i in range(len(nv)) for j in range(i+1,len(nv))]    # add the intercept    return [sum(x) for x in zip(c, b)]# This replicates clf.predict(X)def predict(params, sv, nv, a, b, cs, X):    ''' params = model parameters        sv = support vectors        nv = # of support vectors per class        a  = dual coefficients        b  = intercepts         cs = list of class names        X  = feature to predict           '''    decision = decision_function(params, sv, nv, a, b, X)    votes = [(i if decision[p] > 0 else j) for p,(i,j) in enumerate((i,j)                                            for i in range(len(cs))                                           for j in range(i+1,len(cs)))]    return cs[max(set(votes), key=votes.count)]

最近两周花了很长时间摸索，主要原因是因为不清楚为什么sv对应decision_function返回的distance很少有是1的距离，究其原因还是对模型理解不够透彻，忽略了overlap时，outliner的data也是sv.

希望对大家有一点帮助。
以上