Python下载网络文本数据到本地内存的四种实现方法示例

Posted 2021-01-16 suesun

reference: https://blog.csdn.net/john_bian/article/details/71025372?utm_source=itdadao&utm_medium=referral

import urllib.request
import requests
from io import StringIO
import numpy as np
import pandas as pd
‘‘‘
下载网络文件，并导入CSV文件作为numpy的矩阵
‘‘‘
# 网络数据文件地址
url = "http://archive.ics.uci.edu/ml/machine-learning-databases/pima-indians-diabetes/pima-indians-diabetes.data"
# 方法一
# ========================================================
# 下载文件
r = urllib.request.urlopen(url)
# 导入CSV文件作为numpy的矩阵
dataset = np.loadtxt(r, delimiter=",")
# 方法二
# ========================================================
# 下载文件
r = requests.get(url)
# 导入CSV文件作为numpy的矩阵
dataset = np.loadtxt(StringIO(r.text), delimiter=",") # 此处用到 StringIO ！！！！！！
# 方法三
# ========================================================
#用genfromtxt直接下载网络文件，并将CSV文件导作numpy矩阵。爽！！！！！！！！
dataset = np.genfromtxt(url, delimiter=",")
# 方法四
# ========================================================
# 用pandas.read_csv直接下载网络文件，并将CSV文件导作pandas.DataFrame。
dataset = pd.read_csv(‘http://www-bcf.usc.edu/~gareth/ISL/Advertising.csv‘, index_col=0)
dataset = pd.read_csv(url)
# ========================================================
# separate the data from the target attributes
X = dataset[:,0:7]
y = dataset[:,8]
print(X)
print(y)