http://scikit-learn.org/stable/auto_examples/feature_stacker.html

在许多实际应用中，会有很多方法可以从一个数据集中提取特征。也常常会组合多个方法来获得良好的特征。这个例子说明如何使用FeatureUnion 来结合由PCA 和univariate selection 时的特征。

这个范例的主要目的：

1. 资料集：iris 鸢尾花资料集
2. 特征：鸢尾花特征
3. 预测目标：是那一种鸢尾花
4. 机器学习方法：SVM 支持向量机
5. 探讨重点：特征结合
6. 关键函式： sklearn.pipeline.FeatureUnion

(一)资料汇入及描述

• 首先先汇入iris 鸢尾花资料集，使用from sklearn.datasets import load_iris将资料存入
• 准备X (特征资料) 以及 y (目标资料)

from sklearn.pipeline import Pipeline, FeatureUnion
from sklearn.grid_search import GridSearchCV
from sklearn.svm import SVC
from sklearn.datasets import load_iris
from sklearn.decomposition import PCA
from sklearn.feature_selection import SelectKBest
iris = load_iris()
X, y = iris.data, iris.target

测试资料：

iris为一个dict型别资料。

(二)PCA与SelectKBest

• PCA(n_components = 主要成份数量):Principal Component Analysis(PCA)主成份分析，是一个常用的将资料维度减少的方法。它的原理是找出一个新的座标轴，将资料投影到该轴时，数据的变异量会最大。利用这个方式减少资料维度，又希望能保留住原数据点的特性。

• SelectKBest(score_func , k ): score_func是选择特征值所依据的函式，而K值则是设定要选出多少特征。

# This dataset is way to high-dimensional. Better do PCA:
pca = PCA(n_components=2)
# Maybe some original features where good, too?
selection = SelectKBest(k=1)

(三)FeatureUnionc

• 使用sklearn.pipeline.FeatureUnion合併主成分分析(PCA)和综合筛选(SelectKBest)。
• 最后得到选出的特征

# Build estimator from PCA and Univariate selection:
combined_features = FeatureUnion([("pca", pca), ("univ_select", selection)])
# Use combined features to transform dataset:
X_features = combined_features.fit(X, y).transform(X)

(四)找到最佳的结果

• Scikit-learn的支持向量机分类函式库利用 SVC() 建立运算物件，之后并可以用运算物件内的方法 .fit() 与 .predict() 来做训练与预测。

• 使用GridSearchCV交叉验证，得到由参数网格计算出的分数网格，并找到分数网格中最佳点。最后显示这个点所代表的参数

svm = SVC(kernel="linear")
# Do grid search over k, n_components and C:
pipeline = Pipeline([("features", combined_features), ("svm", svm)])
param_grid = dict(features__pca__n_components=[1, 2, 3],
                  features__univ_select__k=[1, 2],
                  svm__C=[0.1, 1, 10])
grid_search = GridSearchCV(pipeline, param_grid=param_grid, verbose=10)
grid_search.fit(X, y)
print(grid_search.best_estimator_)

结果显示
``` Fitting 3 folds for each of 18 candidates, totalling 54 fits
[CV] featuresuniv_selectk=1, featurespcan_components=1, svmC=0.1
[CV] featuresuniv_selectk=1, featurespcan_components=1, svmC=0.1, score=0.960784 - 0.0s

## (五)完整程式码
Python source code: feature_stacker.py
http://scikit-learn.org/stable/auto_examples/feature_stacker.html
```python
# Author: Andreas Mueller <amueller@ais.uni-bonn.de>
#
# License: BSD 3 clause
from sklearn.pipeline import Pipeline, FeatureUnion
from sklearn.grid_search import GridSearchCV
from sklearn.svm import SVC
from sklearn.datasets import load_iris
from sklearn.decomposition import PCA
from sklearn.feature_selection import SelectKBest
iris = load_iris()
X, y = iris.data, iris.target
# This dataset is way to high-dimensional. Better do PCA:
pca = PCA(n_components=2)
# Maybe some original features where good, too?
selection = SelectKBest(k=1)
# Build estimator from PCA and Univariate selection:
combined_features = FeatureUnion([("pca", pca), ("univ_select", selection)])
# Use combined features to transform dataset:
X_features = combined_features.fit(X, y).transform(X)
svm = SVC(kernel="linear")
# Do grid search over k, n_components and C:
pipeline = Pipeline([("features", combined_features), ("svm", svm)])
param_grid = dict(features__pca__n_components=[1, 2, 3],
                  features__univ_select__k=[1, 2],
                  svm__C=[0.1, 1, 10])
grid_search = GridSearchCV(pipeline, param_grid=param_grid, verbose=10)
grid_search.fit(X, y)
print(grid_search.best_estimator_)