![](https://img-blog.csdnimg.cn/20201014180756923.png?x-oss-process=image/resize,m_fixed,h_64,w_64)
python
sinat_36190649
当老师
展开
-
运用python进行相关性统计检验
本文主要记录了python对数据进行相关性统计检验的实现方法。主要是:1、连续变量VS连续变量:皮尔逊相关系数2、分类变量VS分类变量:卡方检验3、连续变量VS多分类变量:F检验4、连续变量VS二分类变量:若连续变量满足正态分布,则用t检验,否则应用曼-惠特尼U检验主要的python脚本如下:from sklearn.feature_selection import SelectKBest,chi2from scipy import statsfrom scipy.sta原创 2020-12-14 15:56:27 · 2602 阅读 · 0 评论 -
python实现stacking
stacking是模型融合的一种方法,参考https://www.sohu.com/a/302683886_787107def evaluation(model,test_x,test_y,model_type = 'other_model'): predict = model.predict(test_x) if model_type == 'lgb': y_score = np.array(predict) predict = [1 if x >原创 2020-09-07 19:23:48 · 558 阅读 · 0 评论 -
python得到地址的经纬度,并将经纬度转成详细地址
import pandas as pdimport jsonfrom urllib.request import urlopen, quoteimport requestsimport reimport pandas as pdfrom geopy.geocoders import Nominatim,GoogleV3from geopy.exc import GeocoderTimedOutfirst_aid = pd.read_pickle('t_first_aid.pkl')# .原创 2020-08-14 15:52:00 · 1848 阅读 · 0 评论 -
python技巧积累
#从dataframe中将两列符合条件的行取出来def info_duplicates(base_info): base_info['miss'] = base_info.isnull().sum(axis = 1) tmp_miss_df = base_info['miss'].groupby(base_info['patient_id']).min() zip_list = list(zip(tmp_miss_df.index,list(tmp_miss_df))) .原创 2020-08-11 18:28:00 · 177 阅读 · 0 评论 -
python连接mysql读取数据(总读取与分批读取)
import pymysqlimport pandas as pdimport gcimport timeimport threadingclass Sql_df(object): def __init__(self,input_db): self.host = 'ip_xxx' self.port = 3306 self.username = 'root' self.password = 'xxx' se.原创 2020-08-11 18:25:21 · 1619 阅读 · 0 评论 -
python学习—正则表达式
1.字符串中startswith的用法txt = ['mooc yuwen','mooc shuxue mooc','mooc yingyu','wuli','huaxue','shengwu']for item in txt: if item.startswith('mooc'):#字符串中的startswith的用法 print(item)for item in t...原创 2018-06-07 11:36:28 · 249 阅读 · 0 评论 -
python学习—数据处理
import pandas as pdimport numpy as npdf = pd.read_csv('C:\\Users\\len\\Desktop\\statistic_race\\pinganchanxian\\user_data.csv')label = df.labeldf = df.drop(['user_id','label'],axis = 1)#删除列,axis=1...原创 2018-06-21 11:18:08 · 366 阅读 · 0 评论 -
python学习—collections模块学习
tuple的功能:1.tuple是不可变的,具有迭代(iterable)性2.拆包用法:name,age,height = ('kang',22,179)3.元组不是绝对不可变的,可对tuple中的列表等进行操作4.tuple的immutable的重要性,即不可变性地重要性,可将元组整体作为字典的键from collections import namedtuple*other,address ...原创 2018-06-21 11:08:04 · 151 阅读 · 0 评论 -
python学习—scipy知识点
1.1 integral积分运算import numpy as npfrom scipy.integrate import quad,dblquad,nquad #quad是一元积分,dblquad是二元积分,nquad表示n维的积分print(quad(lambda x:np.exp(-x),0,np.inf))#在进行一元积分时此处0是x的下界,np.inf是x的上界print(dblq...原创 2018-06-21 11:00:54 · 796 阅读 · 0 评论 -
python学习—装饰器
1.闭包的相关知识点def func_150(val): passline = 90 if val >= passline: print('pass') else: print('failed')def func_100(val): passline = 60 if val >= passline: ...原创 2018-06-21 10:52:30 · 124 阅读 · 0 评论