sklearn特征提取

from sklearn.datasets import load_iris
from sklearn.model_selection import train_test_split #集合划分
from sklearn.feature_extraction import DictVectorizer #字典特征抽取
from sklearn.feature_extraction.text import CountVectorizer #根据词频文本特征提取
from sklearn.feature_extraction.text import TfidfVectorizer #根据tf-idf文本特征提取

import numpy as np
import pandas as pd
import jieba #分词
def datasets_demo():
    iris=load_iris()
    x_train,x_test,y_train,y_test= train_test_split(iris.data,iris.target,test_size=0.2,random_state=22)
    print(x_train,x_train.shape)
    #    print(iris)
  #  print(iris['DESCR'])
    return None

def dic_demo():
    data=[{"city":'北京','temp':100},
          {"city":'上海','temp':100},
          {"city":'南京','temp':100}]
    transfer=DictVectorizer(sparse=False)
    data_new=transfer.fit_transform(data)
    print(data_new)
    print(transfer.get_feature_names())
    return None
def text_demo():
    text=["life is short,i like like python","life is too long,i dislike python"]
    transfer=CountVectorizer()
    text_new=transfer.fit_transform(text)
    print(transfer.get_feature_names())
    print(text_new.toarray())
def cut_word(text):
    a=" ".join(list(jieba.cut(text)))
    print(type(a))
    return a
def count_chinese_demo2():
    data=["一种还是一种今天很残酷，明天更残酷，后天很美好",
          "我们看到的从很远星系来的光是在几百万年之前发出的",
          "如果只从一种方式了解某样事物，你就不会真正了解它"]
    text=[]
    #分词
    for sent in data:
        str=" ".join(jieba.cut(sent))
        text.append(str)
    #字符统计
    transfer=CountVectorizer(stop_words=["一种","不会"])
    text_new=transfer.fit_transform(text)
    print(transfer.get_feature_names())
    print(text_new.toarray())

    print(text)
def tfidf_demo():
    data = ["一种还是一种今天很残酷，明天更残酷，后天很美好",
            "我们看到的从很远星系来的光是在几百万年之前发出的",
            "如果只从一种方式了解某样事物，你就不会真正了解它"]
    text = []
    # 分词
    for sent in data:
        str = " ".join(jieba.cut(sent))
        text.append(str)
    # 字符统计
    transfer = TfidfVectorizer(stop_words=["一种", "不会"])
    text_new = transfer.fit_transform(text)
    print(transfer.get_feature_names())
    print(text_new.toarray())

if __name__ =='__main__':
    #datasets_demo()
    #dic_demo()
    #text_demo()
    #print(cut_word("我爱北京天安门"))
    #count_chinese_demo2()
    #tfidf_demo()