引言
在编程中经常会遇到图片等数据集将图片等数据以URL形式存储在txt文档中，为便于后续的分析，需要将其下载下来，并按照文件夹分类存储。本文以Github中Alexander Kim提供的图片分类数据集为例，下载其提供的图片样本并分类保存
Python 3.6.5，Anaconda， VSCode
1. 下载数据集文件
建立项目文件夹，下载上述Github项目中的raw_data文件夹，并保存至项目目录中。
2. 获取样本文件位置
编写get_doc_path.py，根据根目录位置，获取目录及其子目录所有数据集文件
import os
def get_file(root_path, all_files={}):
'''
递归函数，遍历该文档目录和子目录下的所有文件，获取其path
'''
files = os.listdir(root_path)
for file in files:
if not os.path.isdir(root_path + '/' + file): # not a dir
all_files[file] = root_path + '/' + file
else: # is a dir
get_file((root_path+'/'+file), all_files)
return all_files
if __name__ == '__main__':
path = './raw_data'
print(get_file(path))
3. 下载文件
3.1 读取url列表并
for filename, path in paths.items():
print('reading file: {}'.format(filename))
with open(path, 'r') as f:
lines = f.readlines()
url_list = []
for line in lines:
url_list.append(line.strip('\n'))
print(url_list)
3.2 创建文件夹
foldername = "./picture_get_by_url/pic_download/{}".format(filename.split('.')[0])
if not os.path.exists(folder_path):
print("Selected folder not exist, try to create it.")
os.makedirs(folder_path)
3.3 下载图片
def get_pic_by_url(folder_path, lists):
if not os.path.exists(folder_path):
print("Selected folder not exist, try to create it.")
os.makedirs(folder_path)
for url in lists:
print("Try downloading file: {}".format(url))
filename = url.split('/')[-1]
filepath = folder_path + '/' + filename
if os.path.exists(filepath):
print("File have already exist. skip")
else:
try:
urllib.request.urlretrieve(url, filename=filepath)
except Exception as e:
print("Error occurred when downloading file, error message:")
print(e)
4. 完整源码
4.1 get_doc_path.py
import os
def get_file(root_path, all_files={}):
'''
递归函数，遍历该文档目录和子目录下的所有文件，获取其path
'''
files = os.listdir(root_path)
for file in files:
if not os.path.isdir(root_path + '/' + file): # not a dir
all_files[file] = root_path + '/' + file
else: # is a dir
get_file((root_path+'/'+file), all_files)
return all_files
if __name__ == '__main__':
path = './raw_data'
print(get_file(path))
4.2 get_pic.py
import get_doc_path
import os
import urllib.request
def get_pic_by_url(folder_path, lists):
if not os.path.exists(folder_path):
print("Selected folder not exist, try to create it.")
os.makedirs(folder_path)
for url in lists:
print("Try downloading file: {}".format(url))
filename = url.split('/')[-1]
filepath = folder_path + '/' + filename
if os.path.exists(filepath):
print("File have already exist. skip")
else:
try:
urllib.request.urlretrieve(url, filename=filepath)
except Exception as e:
print("Error occurred when downloading file, error message:")
print(e)
if __name__ == "__main__":
root_path = './picture_get_by_url/raw_data'
paths = get_doc_path.get_file(root_path)
print(paths)
for filename, path in paths.items():
print('reading file: {}'.format(filename))
with open(path, 'r') as f:
lines = f.readlines()
url_list = []
for line in lines:
url_list.append(line.strip('\n'))
foldername = "./picture_get_by_url/pic_download/{}".format(filename.split('.')[0])
get_pic_by_url(foldername, url_list)
4.3 运行结果
执行get_pic.py
当程序意外停止或再次执行时，程序会自动跳过文件夹中已下载的文件，继续下载未下载的内容
{‘urls_drawings.txt': ‘./picture_get_by_url/raw_data/drawings/urls_drawings.txt', ‘urls_hentai.txt': ‘./picture_get_by_url/raw_data/hentai/urls_hentai.txt', ‘urls_neutral.txt': ‘./picture_get_by_url/raw_data/neutral/urls_neutral.txt', ‘urls_porn.txt': ‘./picture_get_by_url/raw_data/porn/urls_porn.txt', ‘urls_sexy.txt': ‘./picture_get_by_url/raw_data/sexy/urls_sexy.txt'}
reading file: urls_drawings.txt
Try downloading file: http://41.media.tumblr.com/xxxxxx.jpg
Try downloading file: http://41.media.tumblr.com/xxxxxx.jpg
Try downloading file: http://ak1.polyvoreimg.com/cgi/img-thing/size/l/tid/xxxxxx.jpg
Error occurred when downloading file, error message:
HTTP Error 502: No data received from server or forwarder
Try downloading file: http://akicocotte.weblike.jp/gaugau/xxxxxx.jpg
Try downloading file: http://animewriter.files.wordpress.com/2009/01/nagisa-xxxxxx-xxxxxx.jpg
Try downloading file: http://cdn.awwni.me/xxxxxx.jpg
Try downloading file: http://cdn.awwni.me/xxxxxx.jpg
Try downloading file: http://cdn.awwni.me/xxxxxx.jpg
Try downloading file: http://cdn.awwni.me/xxxxxx.jpg
Try downloading file: http://cdn.awwni.me/xxxxxx.jpg
Try downloading file: http://cdn.awwni.me/xxxxxx.jpg
Try downloading file: http://cdn.awwni.me/xxxxxx.jpg
Try downloading file: http://cdn.awwni.me/xxxxxx.jpg
Try downloading file: http://cdn.awwni.me/xxxxxx.jpg
Try downloading file: http://cdn.awwni.me/xxxxxx.jpg
Try downloading file: http://cdn.awwni.me/xxxxxx.jpg
Try downloading file: http://cdn.awwni.me/xxxxxx.jpg
Try downloading file: http://cdn.awwni.me/xxxxxx.jpg
Try downloading file: http://cdn.awwni.me/xxxxxx.jpg
Try downloading file: http://cdn.awwni.me/xxxxxx.jpg
Try downloading file: http://cdn.awwni.me/xxxxxx.jpg
Try downloading file: http://cdn.awwni.me/xxxxxx.jpg
后注：由于样本数据集内容的问题，上述地址以xxxxx代替具体地址，案例项目也已经失效，但是方法仍然可以借鉴
20.9.23更新：数据集地址：https://github.com/ZQ-Qi/nsfw_data_scrapper，单纯为了学习和实践本文代码的可以下载该数据集进行尝试