模型融合

参考：台大机器学习技法  http://blog.csdn.net/lho2010/article/details/42927287

          stacking&blending  http://heamy.readthedocs.io/en/latest/usage.html

1.blending

比如数据分成train和test，对于model_i（比如xgboost） ，即对所有的数据训练模型model_i,预测test数据生成预测向量v_i, 然后对train做CV fold=5，  然后对其他4份做训练数据，另外一份作为val数据，得出模型model_i_j，然后对val预测生成向量t_i_j, 然后将5分向量concat生成t_i,这是对应t_i与v_i对应，  每个模型都能生成这样一组向量，然后在顶层的模型比如LR或者线性对t向量进行训练，生成blender模型对v向量进行预测

也就是需要生成如下的一个表，训练集数据为把数据切分交叉生成，测试集为训练数据全部训练对测试集预测生成

idmodel_1model_2model_3model_4label10.10.20.140.15020.20.220.180.3130.80.70.880.6140.30.30.20.22050.50.30.60.51
blending 的优点是：比stacking简单，不会造成数据穿越，generalizers和stackers使用不同的数据，可以随时添加其他模型到blender中。

与stacking的区别是：

 stacking在预测 测试集上时直接基于训练数据的

 blender在预测 测试集上每次cv的子集都会预测下预测集， n次cv取平均

Blending：用不相交的数据训练不同的 Base Model，将它们的输出取（加权）平均。

Stacking：划分训练数据集为两个不相交的集合，在第一个集合上训练多个学习器，在第二个集合上测试这几个学习器，把第三步得到的预测结果作为输入，把正确的回应作为输出，训练一个高层学习器。

from __future__ import divisionimport numpy as npimport load_datafrom sklearn.cross_validation import StratifiedKFoldfrom sklearn.ensemble import RandomForestClassifier, ExtraTreesClassifier, GradientBoostingClassifierfrom sklearn.linear_model import LogisticRegressionfrom utility import *from evaluator import *def logloss(attempt, actual, epsilon=1.0e-15):    """Logloss, i.e. the score of the bioresponse competition.    """    attempt = np.clip(attempt, epsilon, 1.0-epsilon)    return - np.mean(actual * np.log(attempt) + (1.0 - actual) * np.log(1.0 - attempt))if __name__ == '__main__':    np.random.seed(0) # seed to shuffle the train set    # n_folds = 10    n_folds = 5    verbose = True    shuffle = False    # X, y, X_submission = load_data.load()    train_x_id, train_x, train_y = preprocess_train_input()    val_x_id, val_x, val_y = preprocess_val_input()    X = train_x    y = train_y    X_submission = val_x    X_submission_y = val_y    if shuffle:        idx = np.random.permutation(y.size)        X = X[idx]        y = y[idx]    skf = list(StratifiedKFold(y, n_folds))    clfs = [RandomForestClassifier(n_estimators=100, n_jobs=-1, criterion='gini'),            RandomForestClassifier(n_estimators=100, n_jobs=-1, criterion='entropy'),            ExtraTreesClassifier(n_estimators=100, n_jobs=-1, criterion='gini'),            ExtraTreesClassifier(n_estimators=100, n_jobs=-1, criterion='entropy'),            GradientBoostingClassifier(learning_rate=0.05, subsample=0.5, max_depth=6, n_estimators=50)]    print "Creating train and test sets for blending."        dataset_blend_train = np.zeros((X.shape[0], len(clfs)))    dataset_blend_test = np.zeros((X_submission.shape[0], len(clfs)))        for j, clf in enumerate(clfs):        print j, clf        dataset_blend_test_j = np.zeros((X_submission.shape[0], len(skf)))        for i, (train, test) in enumerate(skf):            print "Fold", i            X_train = X[train]            y_train = y[train]            X_test = X[test]            y_test = y[test]            clf.fit(X_train, y_train)            y_submission = clf.predict_proba(X_test)[:,1]            dataset_blend_train[test, j] = y_submission            dataset_blend_test_j[:, i] = clf.predict_proba(X_submission)[:,1]        dataset_blend_test[:,j] = dataset_blend_test_j.mean(1)        print("val auc Score: %0.5f" % (evaluate2(dataset_blend_test[:,j], X_submission_y)))    print    print "Blending."    # clf = LogisticRegression()    clf = GradientBoostingClassifier(learning_rate=0.02, subsample=0.5, max_depth=6, n_estimators=100)    clf.fit(dataset_blend_train, y)    y_submission = clf.predict_proba(dataset_blend_test)[:,1]    print "Linear stretch of predictions to [0,1]"    y_submission = (y_submission - y_submission.min()) / (y_submission.max() - y_submission.min())    print "blend result"    print("val auc Score: %0.5f" % (evaluate2(y_submission, X_submission_y)))    print "Saving Results."    np.savetxt(fname='blend_result.csv', X=y_submission, fmt='%0.9f')

2.rank_avg

这种融合方法适合排序评估指标，比如auc之类的

其中weight_i为该模型权重，权重为1表示平均融合

rank_i表示样本的升序排名 ，也就是越靠前的样本融合后也越靠前

能较快的利用排名融合多个模型之间的差异，而不用去加权样本的概率值融合

3.weighted

加权融合，给模型一个权重weight，然后加权得到最终结果

weight为1时为均值融合，result_i为模型i的输出

4.bagging

从特征，参数，样本的多样性差异性来做多模型融合，参考随机森林