使用seaborn的load_dataset()函数触发URLError问题求助

问题详情 #

运行以下代码：

import seaborn as sns
df = sns.load_dataset('iris')

出现TimeoutError及后续URLError，完整错误栈如下：

---------------------------------------------------------------------------
TimeoutError                              Traceback (most recent call last)
File c:\Users\Argish\AppData\Local\Programs\Python\Python38\lib\urllib\request.py:1354, in AbstractHTTPHandler.do_open(self, http_class, req, **http_conn_args)
   1353 try:
-> 1354     h.request(req.get_method(), req.selector, req.data, headers,
   1355               encode_chunked=req.has_header('Transfer-encoding'))
   1356 except OSError as err: # timeout error

File c:\Users\Argish\AppData\Local\Programs\Python\Python38\lib\http\client.py:1252, in HTTPConnection.request(self, method, url, body, headers, encode_chunked)
   1251 """Send a complete request to the server."""
-> 1252 self._send_request(method, url, body, headers, encode_chunked)

File c:\Users\Argish\AppData\Local\Programs\Python\Python38\lib\http\client.py:1298, in HTTPConnection._send_request(self, method, url, body, headers, encode_chunked)
   1297     body = _encode(body, 'body')
-> 1298 self.endheaders(body, encode_chunked=encode_chunked)

File c:\Users\Argish\AppData\Local\Programs\Python\Python38\lib\http\client.py:1247, in HTTPConnection.endheaders(self, message_body, encode_chunked)
   1246     raise CannotSendHeader()
-> 1247 self._send_output(message_body, encode_chunked=encode_chunked)

File c:\Users\Argish\AppData\Local\Programs\Python\Python38\lib\http\client.py:1007, in HTTPConnection._send_output(self, message_body, encode_chunked)
   1006 del self._buffer[:]
-> 1007 self.send(msg)
   1009 if message_body is not None:
   1010 
   1011     # create a consistent interface to message_body

File c:\Users\Argish\AppData\Local\Programs\Python\Python38\lib\http\client.py:947, in HTTPConnection.send(self, data)
    946 if self.auto_open:
-> 947     self.connect()
    948 else:

File c:\Users\Argish\AppData\Local\Programs\Python\Python38\lib\http\client.py:1414, in HTTPSConnection.connect(self)
   1412 "Connect to a host on a given (SSL) port."
-> 1414 super().connect()
   1416 if self._tunnel_host:

File c:\Users\Argish\AppData\Local\Programs\Python\Python38\lib\http\client.py:918, in HTTPConnection.connect(self)
    917 """Connect to the host and port specified in __init__."""
-> 918 self.sock = self._create_connection(
    919     (self.host,self.port), self.timeout, self.source_address)
    920 self.sock.setsockopt(socket.IPPROTO_TCP, socket.TCP_NODELAY, 1)

File c:\Users\Argish\AppData\Local\Programs\Python\Python38\lib\socket.py:808, in create_connection(address, timeout, source_address)
    807 try:
-> 808     raise err
    809 finally:
    810     # Break explicitly a reference cycle

File c:\Users\Argish\AppData\Local\Programs\Python\Python38\lib\socket.py:796, in create_connection(address, timeout, source_address)
    795     sock.bind(source_address)
-> 796 sock.connect(sa)
    797 # Break explicitly a reference cycle

TimeoutError: [WinError 10060] A connection attempt failed because the connected party did not properly respond after a period of time, or established connection failed because connected host has failed to respond

During handling of the above exception, another exception occurred:

URLError                                  Traceback (most recent call last)
Cell In[14], line 1
----> 1 df = sns.load_dataset('iris')

File c:\Users\Argish\AppData\Local\Programs\Python\Python38\lib\site-packages\seaborn\utils.py:588, in load_dataset(name, cache, data_home, **kws)
    586         if name not in get_dataset_names():
    587             raise ValueError(f"'{name}' is not one of the example datasets.")
-> 588         urlretrieve(url, cache_path)
    589     full_path = cache_path
    590 else:

File c:\Users\Argish\AppData\Local\Programs\Python\Python38\lib\urllib\request.py:247, in urlretrieve(url, filename, reporthook, data)
    230 """
    231 Retrieve a URL into a temporary location on disk.
    232 
   (...)
    243 data file as well as the resulting HTTPMessage object.
    244 """
    245 url_type, path = _splittype(url)
-> 247 with contextlib.closing(urlopen(url, data)) as fp:
    248     headers = fp.info()
    250     # Just return the local path and the "headers" for file://
    251     # URLs. No sense in performing a copy unless requested.

File c:\Users\Argish\AppData\Local\Programs\Python\Python38\lib\urllib\request.py:222, in urlopen(url, data, timeout, cafile, capath, cadefault, context)
    220 else:
    221     opener = _opener
-> 222 return opener.open(url, data, timeout)

File c:\Users\Argish\AppData\Local\Programs\Python\Python38\lib\urllib\request.py:525, in OpenerDirector.open(self, fullurl, data, timeout)
    522     req = meth(req)
    524 sys.audit('urllib.Request', req.full_url, req.data, req.headers, req.get_method())
-> 525 response = self._open(req, data)
    527 # post-process response
    528 meth_name = protocol+"_response"

File c:\Users\Argish\AppData\Local\Programs\Python\Python38\lib\urllib\request.py:542, in OpenerDirector._open(self, req, data)
    539     return result
    541 protocol = req.type
-> 542 result = self._call_chain(self.handle_open, protocol, protocol +
    543                           '_open', req)
    544 if result:
    545     return result

File c:\Users\Argish\AppData\Local\Programs\Python\Python38\lib\urllib\request.py:502, in OpenerDirector._call_chain(self, chain, kind, meth_name, *args)
    500 for handler in handlers:
    501     func = getattr(handler, meth_name)
-> 502     result = func(*args)
    503     if result is not None:
    504         return result

File c:\Users\Argish\AppData\Local\Programs\Python\Python38\lib\urllib\request.py:1397, in HTTPSHandler.https_open(self, req)
   1396 def https_open(self, req):
-> 1397     return self.do_open(http.client.HTTPSConnection, req,
   1398         context=self._context, check_hostname=self._check_hostname)

File c:\Users\Argish\AppData\Local\Programs\Python\Python38\lib\urllib\request.py:1357, in AbstractHTTPHandler.do_open(self, http_class, req, **http_conn_args)
   1354         h.request(req.get_method(), req.selector, req.data, headers,
   1355                   encode_chunked=req.has_header('Transfer-encoding'))
   1356     except OSError as err: # timeout error
-> 1357         raise URLError(err)
   1358     r = h.getresponse()
   1359 except:

URLError: 

已尝试卸载重装seaborn、排查防火墙设置（未阻止Python联网），均无效；切换到Python 3.10.6后代码可正常运行。

解决方法 #

1. 手动填充本地数据集缓存 #

seaborn会优先读取本地缓存的数据集，无需联网：

• 手动创建缓存目录：C:\Users\Argish\AppData\Local\seaborn\seaborn-data
• 从Python 3.10环境的对应缓存目录中复制iris.csv到上述路径（3.10的缓存路径一般是C:\Users\Argish\AppData\Local\seaborn\seaborn-data）
• 重新运行代码，将自动读取本地文件

2. 修复Python 3.8的SSL/网络组件 #

Python 3.8.10的urllib或SSL库可能存在兼容性问题，导致无法连接数据源服务器：

• 升级pip：

python -m pip install --upgrade pip

• 更新证书库：

python -m pip install --upgrade certifi

• 临时关闭SSL证书验证（仅用于测试，不推荐生产环境）：

import ssl
import seaborn as sns
ssl._create_default_https_context = ssl._create_unverified_context
df = sns.load_dataset('iris')

3. 绕开seaborn加载函数，直接用pandas读取 #

如果上述方法都不生效，直接读取本地的iris.csv文件：

import pandas as pd
df = pd.read_csv('C:/Users/Argish/AppData/Local/seaborn/seaborn-data/iris.csv')

内容的提问来源于stack exchange，提问作者Argish Abhangi