首页 > 解决方案 > 文件 csv:如何使用 python 根据其他列值计算值的出现次数?

问题描述

我有一个 .csv 文件(600 行),其中包含一些字段:提交 ID、气味类型等。

我会计算每个提交 ID 的每种气味的出现次数。

我会输出的示例:

   commit dfbu3u4498fbbefi: [dense structure :1, cyclic dependency:4, unstable dependency: 67, feature concentration: 6, god component: 8]
  commit  bifueifyuwefbvwr: [dense structure :34, cyclic dependency:43, unstable dependency: 97, feature concentration: 43, god component: 10]

我试过这个,但我认为我需要另一个循环(也许?)对不起,我以前从未使用过 Python

import csv
import collections

smell = collections.Counter()


with open('Ref.csv') as file:
    reader = csv.reader(file, delimiter=';')

    for row in reader:

        smell[row[0]] += 1

print (smell.most_common(5))

OUTPUT:

[('9b0dd5dc979bd490ae34f6d790c466b47c84c920', 96), ('6431099fe7d5d90da678a78051f12894da82c68d', 96), ('44fdfa7ea93c15bb116a25e0675d98469deafaa6', 96), ('b2c40612a2c60685555f35af71f5801391a58b4b', 96), ('aa6cbb78cca17a9de339b2d060c00352e8beedde', 96)]

or if i change row index to 2 i got

[('Unstable Dependency', 315), ('Feature Concentration', 238), ('God Component', 84), ('Cyclic Dependency', 28), ('Dense Structure', 7)]

标签: pythonpandasloopscsv

解决方案


你可以用pandas它来做到这一点:

import pandas as pd

# Dataframe definition
df = pd.read_csv('Ref.csv', sep=';')

# Group and get the count values.

df_grouped = df.groupby(by=['commit', 'smell']).size()

df_grouped现在是 a pandas.series,如果你想让它dataframe再次成为 a ,你应该这样做:

df_grouped = df_grouped.reset_index()
df_grouped = df_grouped.rename(columns={0: "counts"})

我强烈建议您查看文档:https ://pandas.pydata.org/pandas-docs/stable/index.html


推荐阅读