When using Python for data analysis, two methods for CSV file import | numpy | pandas

Hello everyone, today I will share with you two ways to import CSV files when we use Python for data analysis.

Introduction to iris data set CSV

Dataset: iris.csv

Its structure is as follows. It is a matrix with 150 rows and 5 columns. The first four columns belong to the input features, and the last column belongs to the classification label.

numpy import CSV

import numpy as np
filename='D:/数学建模2022/算法/SVM/iris.csv'
with open(filename,'rt') as raw_data: #一个文件如果要导入python进行后续操作，必须先打开它，并返回一个文件对象
    iris=np.loadtxt(raw_data,delimiter=',')
    print(iris)
    print(iris.shape)

Partial results:

pandas import CSV (my favorite for this)

Pandas has the advantage that you can name the columns

import pandas as pd
file_name='D:/数学建模2022/算法/SVM/iris.csv'
names=['separ-length','separ-width','petal_length','petal_width','class']
data=pd.read_csv(file_name,names=names)
print(data)
print(data.shape)

Well, that's the end of today's sharing, see you next time!