Python-科学计算-pandas-23-按列去重

系统：Windows 10
编辑器：JetBrains PyCharm Community Edition 2018.2.2 x64
pandas：1.1.5

• 这个系列讲讲Python的科学计算及可视化
• 今天讲讲pandas模块
• 将df按某列进行去重

Part 1：场景描述

1. 已知df1，包括6列，"time", "pos", "value1", "value2", "value3", "value4
2. 有两个需求：根据pos列，去除重复记录；根据pos和value1列，去除重复记录，即要求这两列都相等时去重

df_1

Part 2：根据pos列去重

import pandas as pd

dict_1 = {"time": ["2019-11-02", "2019-11-03", "2019-11-04", "2019-11-05",
                   "2019-12-02", "2019-12-03", "2019-12-04", "2019-12-05"],
          "pos": ["A", "A", "C", "D", "E", "E", "G", "H"],
          "value1": [20, 20, 30, 40, 50, 60, 70, 80],
          "value2": [100, 200, 300, 400, 500, 600, 700, 800],
          "value3": [50, 20, 30, 90, 50, 60, 80, 80],
          "value4": ['21W12', '21W10', '21W01', '21W05', '21W06', '21W36', '21W21', '21W23']}

df_1 = pd.DataFrame(dict_1, columns=["time", "pos", "value1", "value2", "value3", "value4"])
print("\n", "df_1", "\n", df_1, "\n")

df_2 = df_1

df_2.drop_duplicates(subset=["pos"], keep="first", inplace=True)
print("\n", "df_2", "\n", df_2, "\n")

print("\n", "df_1", "\n", df_1, "\n")

代码截图

执行结果

Part 3：根据pos和value1列去重

import pandas as pd

dict_1 = {"time": ["2019-11-02", "2019-11-03", "2019-11-04", "2019-11-05",
                   "2019-12-02", "2019-12-03", "2019-12-04", "2019-12-05"],
          "pos": ["A", "A", "C", "D", "E", "E", "G", "H"],
          "value1": [20, 20, 30, 40, 50, 60, 70, 80],
          "value2": [100, 200, 300, 400, 500, 600, 700, 800],
          "value3": [50, 20, 30, 90, 50, 60, 80, 80],
          "value4": ['21W12', '21W10', '21W01', '21W05', '21W06', '21W36', '21W21', '21W23']}

df_1 = pd.DataFrame(dict_1, columns=["time", "pos", "value1", "value2", "value3", "value4"])
print("\n", "df_1", "\n", df_1, "\n")

df_3 = df_1

df_3.drop_duplicates(subset=["pos", "value1"], keep="first", inplace=True)
print("\n", "df_3", "\n", df_3, "\n")

print("\n", "df_1", "\n", df_1, "\n")

代码截图

执行结果

Part 4：部分代码解读

1. df_2.drop_duplicates(subset=["pos"], keep="first", inplace=True)，subset对应列表取值去重参考列，若列表元素大于1个，要求同时满足多列对应记录相同才能去重。keep="first"表示去重后，保留第1个记录
2. df_2=df_1后对，df_2进行去重后，df_1同时发生了变化，表明两个变量对应的地址应该是同一区域

本文为原创作品，欢迎分享朋友圈

长按图片识别二维码，关注本公众号
Python 优雅 帅气