pandas使用
Posted zsmblog
tags:
篇首语：本文由小常识网(cha138.com)小编为大家整理，主要介绍了pandas使用相关的知识，希望对你有一定的参考价值。
import pandas as pd import numpy as np # 读取数据，并以表格的形式显示 df1 = pd.DataFrame(pd.read_excel(‘a.xlsx‘)) # 判断是否为空，每个表格都会去判断，并返回True/False a1 = df1.isnull().values """ [[False False False False False False False False False] [False False True False False False False False False] [False False False False False False False False False] [False False False False False False False False False] [False False False False False True False False False] [False False False False False False False False False] [False False False False False False False False False]] """ # 判断是否为非空，每个表格都会去判断，并返回True/False，会以规则的表格形式返回 a2 = df1.notnull() """ 1 2 3 4 5 6 open_acc weight data 0 True True True True True True True True True 1 True True False True True True True True True 2 True True True True True True True True True 3 True True True True True True True True True 4 True True True True True False True True True 5 True True True True True True True True True 6 True True True True True True True True True """ # 判断列是否重复,比对每一行，并返回True/False b = df1.duplicated() """ 0 False 1 False 2 False 3 False 4 False 5 True 6 False dtype: bool """ # 返回去掉重复行后的表 c = df1.drop_duplicates() """ 1 2 3 4 5 6 open_acc weight data 0 12 12 23 34 56 67 2 12kg 2018/02/26 1 asd asd NaN aas qwe qwe 10 23000g 2018/02/6 2 123 qw q q qwe qwe 7 45000mg 2018/5/26 3 1 2 3 4 5 6 15 123t 2017/9 4 1 2 3 4 5 NaN 20 456t 18/2/2 6 12 123 4 3 er 32 12 34g 2015/3/01 """ # 空值以。。。。方式填充 d1 = df1.fillna(‘\‘) """ 1 2 3 4 5 6 open_acc weight data 0 12 12 23 34 56 67 2 12kg 2018/02/26 1 asd asd aas qwe qwe 10 23000g 2018/02/6 2 123 qw q q qwe qwe 7 45000mg 2018/5/26 3 1 2 3 4 5 6 15 123t 2017/9 4 1 2 3 4 5 20 456t 18/2/2 5 1 2 3 4 5 6 15 123t 2017/9 6 12 123 4 3 er 32 12 34g 2015/3/01 """ # 包含空值的行删除 d2 = df1.dropna() """ 1 2 3 4 5 6 open_acc weight data 0 12 12 23 34 56 67 2 12kg 2018/02/26 2 123 qw q q qwe qwe 7 45000mg 2018/5/26 3 1 2 3 4 5 6 15 123t 2017/9 5 1 2 3 4 5 6 15 123t 2017/9 6 12 123 4 3 er 32 12 34g 2015/3/01 """ # 给表更换新的列名，注意字段列表的长度一定要和列的长度相同 column_names = [‘a‘, ‘b‘, ‘c‘, ‘d‘, ‘e‘, ‘f‘,‘g‘,‘h‘,‘i‘] df = pd.read_excel(‘a.xlsx‘, names=column_names) """ a b c d e f g h i 0 12 12 23 34 56 67 2 12kg 2018/02/26 1 asd asd NaN aas qwe qwe 10 23000g 2018/02/6 2 123 qw q q qwe qwe 7 45000mg 2018/5/26 3 1 2 3 4 5 6 15 123t 2017/9 4 1 2 3 4 5 NaN 20 456t 18/2/2 5 1 2 3 4 5 6 15 123t 2017/9 6 12 123 4 3 er 32 12 34g 2015/3/01 """ # 打印第一列 # print(df1[1]) """ 0 12 1 asd 2 123 3 1 4 1 5 1 6 12 Name: 1, dtype: object """ # 将列中的12，1替换成‘qwe‘ a = df1[1].replace([12, 1], ‘qwe‘) """ 0 qwe 1 qwe 2 123 3 qwe 4 qwe 5 qwe 6 qwe Name: 1, dtype: object """ # 打印数据的前n行，默认为5 # print(df1.head()) # 给数据分组 """ 0-5 A 不包含0 6-10 B 11-15 C 16-20 D """ bins = [0, 5, 10, 15, 20] group_names = [‘A‘, ‘B‘, ‘C‘, ‘D‘] df1[‘categories‘] = pd.cut(df1[‘open_acc‘], bins, labels=group_names) """ 1 2 3 4 ... open_acc weight data categories 0 12 12 23 34 ... 2 12kg 2018/02/26 A 1 1 asd NaN aas ... 10 23000g 2018/02/6 B 2 123 qw q q ... 7 45000mg 2018/5/26 B 3 1 2 3 4 ... 15 123t 2017/9 C 4 1 2 3 4 ... 20 456t 18/2/2 D 5 1 2 3 4 ... 15 123t 2017/9 C 6 12 123 4 3 ... 12 34g 2015/3/01 C """ # 获取到需要换算单位的行 rows_with_lbs = df1[‘weight‘].str.contains(‘mg‘).fillna(False) """ 1 2 3 4 5 6 open_acc weight data categories 2 123 qw q q qwe qwe 7 45000mg 2018/5/26 B """ # 换算单位 for i, lbs_row in df1[rows_with_lbs].iterrows(): # print(i) # 获取到值的行 # print(lbs_row[‘weight‘][-2:]) weight = int(float(lbs_row[‘weight‘][:-2]) / 1000) df1.at[i, ‘weight‘] = ‘{}g‘.format(weight) # print(df1) """ 1 2 3 4 ... open_acc weight data categories 0 12 12 23 34 ... 2 12kg 2018/02/26 A 1 1 asd NaN aas ... 10 23000g 2018/02/6 B 2 123 qw q q ... 7 45g 2018/5/26 B 3 1 2 3 4 ... 15 123t 2017/9 C 4 1 2 3 4 ... 20 456t 18/2/2 D 5 1 2 3 4 ... 15 123t 2017/9 C 6 12 123 4 3 ... 12 34g 2015/3/01 C """ # rows_with_lbs = df1[‘weight‘].str.contains(‘t‘).fillna(False) # for i,lbs_row in df1[rows_with_lbs].iterrows(): # # print(i) # 获取到值的行 # # print(lbs_row[‘weight‘][:-2]) # weight = int(float(lbs_row[‘weight‘][:-2]) * 10000000) # df1.at[i, ‘weight‘] = ‘{}g‘.format(weight) # print(df1) # 换算单位封装 def conversion(DataFrame, column, unit, num, new_unit=‘‘, null=False): """ :param DataFrame: pandas对象 :param column: 字段 :param unit: 当前单位 :param new_unit: 新单位 :param num: 当前单位换算成新单位所需要乘的数 :param null: 是否为空 :return:pandas对象 1 2 3 4 ... open_acc weight data categories 0 12 12 23 34 ... 2 1.20e+04g 2018/02/26 A 1 asd asd NaN aas ... 10 23g 2018/02/6 B 2 123 qw q q ... 7 45g 2018/5/26 B 3 1 2 3 4 ... 15 1.23e+08g 2017/9 C 4 1 2 3 4 ... 20 4.56e+08g 18/2/2 D """ rows_with_lbs = DataFrame[column].str.contains(unit).fillna(null) column_len = len(unit) for i, lbs_row in DataFrame[rows_with_lbs].iterrows(): # print(lbs_row[‘weight‘][-column_len:]) weight = int(float(lbs_row[column][:-column_len]) * num) # 默认4位，超出长度变为科学计数法 if len(str(weight)) > 4: DataFrame.at[i, column] = (‘%0.2e‘ + new_unit) % weight """ 1 2 3 4 ... open_acc weight data categories 0 12 12 23 34 ... 2 1.20e+04g 2018/02/26 A 1 asd asd NaN aas ... 10 2.30e+01g 2018/02/6 B 2 123 qw q q ... 7 4.50e+01g 2018/5/26 B 3 1 2 3 4 ... 15 1.23e+08g 2017/9 C 4 1 2 3 4 ... 20 4.56e+08g 18/2/2 D """ else: DataFrame.at[i, column] = str(weight) + new_unit """ 1 2 3 4 ... open_acc weight data categories 0 12 12 23 34 ... 2 12000g 2018/02/26 A 1 asd asd NaN aas ... 10 23g 2018/02/6 B 2 123 qw q q ... 7 45g 2018/5/26 B 3 1 2 3 4 ... 15 123000000g 2017/9 C 4 1 2 3 4 ... 20 456000000g 18/2/2 D """ return DataFrame df1 = conversion(df1, ‘weight‘, ‘kg‘, 1000, ‘g‘) # kg - g df2 = conversion(df1, ‘weight‘, ‘mg‘, 0.001, ‘g‘) # mg -g df3 = conversion(df2, ‘weight‘, ‘t‘, 1000000, ‘g‘) # t - g print(df3)
以上是关于pandas使用的主要内容，如果未能解决你的问题，请参考以下文章
text [检查特定的数据片段]取自论文但有意思应用。 #python #pandas