python – scikit-learn管道：网格搜索变压器参数以生成数据

import numpy as np
from sklearn.pipeline import Pipeline
from sklearn.grid_search import GridSearchCV
from sklearn.base import TransformerMixin
from sklearn import cluster

class FeatureGenerator(TransformerMixin):

    def __init__(self,num_features=None):
        self.num_features = num_features

    def fit(self,X,y=None,**fit_params):
        return self

    def transform(self,**transform_params):
        return np.array(
            range(self.num_features*self.num_features)
        ).reshape(self.num_features,self.num_features)

    def get_params(self,deep=True):
        return {"num_features": self.num_features}

    def set_params(self,**parameters):
        self.num_features = parameters["num_features"]
        return self

这个变压器在起作用. G.被称为这样：

pipeline = Pipeline([
    ('pick_features',FeatureGenerator(100)),('kmeans',cluster.KMeans())
])

pipeline = pipeline.fit(None)
classes = pipeline.predict(None)
print classes

一旦我尝试对此管道进行网格搜索,它就变得棘手：

parameter_sets = {
    'pick_features__num_features' : [10,20,30],'kmeans__n_clusters' : [2,3,4]
}

pipeline = Pipeline([
    ('pick_features',FeatureGenerator()),cluster.KMeans())
])

g_search_estimator = GridSearchCV(pipeline,parameter_sets)

g_search_estimator.fit(None,None)

网格搜索期望样本和标签作为输入,并且不像管道那样健壮,管道不会抱怨None作为输入参数：

TypeError: Expected sequence or array-like,got <type 'NoneType'>

这是有道理的,因为网格搜索需要将数据集划分为不同的cv分区.

除了上面的例子,我有很多参数,可以在数据集生成步骤中进行调整.因此,我需要一个解决方案,将此步骤包含在我的参数选择交叉验证中.

问题：有没有办法从第一个变换器内部设置GridSearch的X和y？或者解决方案怎么样,用多个不同的数据集(最好是并行)调用GridSearch？或者有没有人试图自定义GridSearchCV或可以指向一些阅读材料？