废话真的一句也不想多说，直接看代码吧！

# -*- coding: utf-8 -*- import numpy from sklearn import metrics from sklearn.svm import LinearSVC from sklearn.naive_bayes import MultinomialNB from sklearn import linear_model from sklearn.datasets import load_iris from sklearn.cross_validation import train_test_split from sklearn.preprocessing import OneHotEncoder, StandardScaler from sklearn import cross_validation from sklearn import preprocessing import scipy as spfrom sklearn.linear_model import LogisticRegressionfrom sklearn.feature_selection import SelectKBest ,chi2import pandas as pdfrom sklearn.preprocessing import OneHotEncoder#import iris_data '''creativeID,userID,positionID,clickTime,conversionTime,connectionType,telecomsOperator,appPlatform,sitesetID,positionType,age,gender,education,marriageStatus,haveBaby,hometown,residence,appID,appCategory,label''' def test(): df = pd.read_table("/var/lib/mysql-files/data1.csv", sep=",") df1 = df[["connectionType","telecomsOperator","appPlatform","sitesetID", "positionType","age","gender","education","marriageStatus", "haveBaby","hometown","residence","appCategory","label"]] print df1["label"].value_counts() N_data = df1[df1["label"]==0] P_data = df1[df1["label"]==1] N_data = N_data.sample(n=P_data.shape[0], frac=None, replace=False, weights=None, random_state=2, axis=0) #print df1.loc[:,"label"]==0 print P_data.shape print N_data.shape data = pd.concat([N_data,P_data]) print data.shape data = data.sample(frac=1).reset_index(drop=True) print data[["label"]] return

补充拓展：pandas实现对dataframe抽样

随机抽样

import pandas as pd#对dataframe随机抽取2000个样本pd.sample(df, n=2000)

分层抽样

利用sklean中的函数灵活进行抽样

from sklearn.model_selection import train_test_split#y是在X中的某一个属性列X_train, X_test, y_train, y_test = train_test_split(X,y, test_size=0.2, stratify=y)

以上这篇python使用pandas抽样训练数据中某个类别实例就是小编分享给大家的全部内容了，希望能给大家一个参考，也希望大家多多支持。