sklearn数据映射之分位数均匀分布映射-Quantile_tranformer

from sklearn.datasets import load_iris
from sklearn.model_selection import train_test_split
from sklearn import preprocessing
import numpy as np
import matplotlib.pyplot as plt

#load the iris dataset
iris = load_iris()
#extrant the features and the target
X, y = iris.data, iris.target
#split the dataset into training dataset and testing dataset
X_train, X_test, y_train, y_test = train_test_split(X, y, random_state = 0)
#construction of quantile_transformer
quantile_transformer = preprocessing.QuantileTransformer(random_state = 0)
#transform the train data by quantile_transformer
X_test_trans = quantile_transformer.fit_transform(X_test)
#print the qth percentitles of the array elements
#print(X_test)
#print(X_test_trans)
#print()
#np.percentile(X_train[:, 0], [0, 25, 50, 75, 100])
#np.percentile(X_test[:, 0], [0, 25, 50, 75, 100])
#np.percentile(X_test_trans[:, 0], [0, 25, 50, 75, 100])

plt.scatter(X_test[:, 0], X_test_trans[:, 0])

<matplotlib.collections.PathCollection at 0x23913bfdd68>

从这个图中我们可以十分清楚的看到实际上，这个映射是将原始的数据集按照累积概率密度函数F(X)
来进行映射的，将每个离散值映射到其累积概率分布，映射之后的X是均匀分布呢？我们知道：

1. $$F(X)=X$$
2. $$F(X)={\int }_0^{X}p(t)\mathrm{d}t$$
   1和2同时对X求导可得
   $$P(X)=1$$
   显然随机变量X是服从均匀分布的，且范围为0-1