Downloading MNIST dataset with torchvision gives HTTP Error 403 #1938

Borda · 2020-03-04T10:36:34Z

🐛 Bug

I'm getting a 403 error when I try to download MNIST dataset with torchvision 0.4.2.

To Reproduce

../.local/lib/python3.6/site-packages/torchvision/datasets/mnist.py:68: in __init__
    self.download()
../.local/lib/python3.6/site-packages/torchvision/datasets/mnist.py:135: in download
    download_and_extract_archive(url, download_root=self.raw_folder, filename=filename)
../.local/lib/python3.6/site-packages/torchvision/datasets/utils.py:248: in download_and_extract_archive
    download_url(url, download_root, filename, md5)
../.local/lib/python3.6/site-packages/torchvision/datasets/utils.py:96: in download_url
    raise e
../.local/lib/python3.6/site-packages/torchvision/datasets/utils.py:84: in download_url
    reporthook=gen_bar_updater()
/usr/local/lib/python3.6/urllib/request.py:248: in urlretrieve
    with contextlib.closing(urlopen(url, data)) as fp:
/usr/local/lib/python3.6/urllib/request.py:223: in urlopen
    return opener.open(url, data, timeout)
/usr/local/lib/python3.6/urllib/request.py:532: in open
    response = meth(req, response)
/usr/local/lib/python3.6/urllib/request.py:642: in http_response
    'http', request, response, code, msg, hdrs)
/usr/local/lib/python3.6/urllib/request.py:570: in error
    return self._call_chain(*args)
/usr/local/lib/python3.6/urllib/request.py:504: in _call_chain
    result = func(*args)
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ 

self = <urllib.request.HTTPDefaultErrorHandler object at 0x7efbf9edaac8>
req = <urllib.request.Request object at 0x7efbf9eda8d0>
fp = <http.client.HTTPResponse object at 0x7efbf9edaf98>, code = 403
msg = 'Forbidden', hdrs = <http.client.HTTPMessage object at 0x7efbf9ea22b0>

    def http_error_default(self, req, fp, code, msg, hdrs):
>       raise HTTPError(req.full_url, code, msg, hdrs, fp)
E       urllib.error.HTTPError: HTTP Error 403: Forbidden

Environment

torch==1.3.1
torchvision==0.4.2

Additional context

https://app.circleci.com/jobs/github/PyTorchLightning/pytorch-lightning/6877

The text was updated successfully, but these errors were encountered:

fmassa · 2020-03-04T12:40:56Z

Thanks for reporting! I can reproduce the issue locally, and downloading from the browser works.

I don't yet know what the root cause is though.

fmassa · 2020-03-04T12:52:54Z

I think we might need to pass header in the download_url function

vision/torchvision/datasets/utils.py

Lines 59 to 100 in c3e2b01

    
           def download_url(url, root, filename=None, md5=None): 
        
               """Download a file from a url and place it in root. 
        
               Args: 
        
                   url (str): URL to download file from 
        
                   root (str): Directory to place downloaded file in 
        
                   filename (str, optional): Name to save the file under. If None, use the basename of the URL 
        
                   md5 (str, optional): MD5 checksum of the download. If None, do not check 
        
               """ 
        
               from six.moves import urllib 
        
               root = os.path.expanduser(root) 
        
               if not filename: 
        
                   filename = os.path.basename(url) 
        
               fpath = os.path.join(root, filename) 
        
               makedir_exist_ok(root) 
        
               # check if file is already present locally 
        
               if check_integrity(fpath, md5): 
        
                   print('Using downloaded and verified file: ' + fpath) 
        
               else:   # download the file 
        
                   try: 
        
                       print('Downloading ' + url + ' to ' + fpath) 
        
                       urllib.request.urlretrieve( 
        
                           url, fpath, 
        
                           reporthook=gen_bar_updater() 
        
                       ) 
        
                   except (urllib.error.URLError, IOError) as e: 
        
                       if url[:5] == 'https': 
        
                           url = url.replace('https:', 'http:') 
        
                           print('Failed download. Trying https -> http instead.' 
        
                                 ' Downloading ' + url + ' to ' + fpath) 
        
                           urllib.request.urlretrieve( 
        
                               url, fpath, 
        
                               reporthook=gen_bar_updater() 
        
                           ) 
        
                       else: 
        
                           raise e 
        
                   # check integrity of downloaded file 
        
                   if not check_integrity(fpath, md5): 
        
                       raise RuntimeError("File not found or corrupted.")

according to https://stackoverflow.com/questions/13303449/urllib2-httperror-http-error-403-forbidden

cc @cpuhrsch @vincentqb @zhangguanheng66 for awareness

soumith · 2020-03-04T12:53:14Z

this is because the download links for mnist at https://github.com/pytorch/vision/blob/master/torchvision/datasets/mnist.py#L33-L36 are hosted on yann.lecun.com and that server has moved under CloudFlare protection.

@fmassa we need to maybe mirror and change the URLs to maybe the PyTorch S3 bucket or something

Borda · 2020-03-04T12:57:34Z

so could we make a hot-fix somehow?

fmassa · 2020-03-04T12:59:27Z

@Borda I haven't tried the current hotfix I mentioned, but I think it might be possible, would you be able to try it and send a PR? Otherwise I'll look into it early next week (I'm working towards ECCV deadline tomorrow)

And I would rather avoid hosting the datasets ourselves, as this would give precedence on us storing the datasets.

eduardo4jesus · 2020-03-04T14:25:51Z

Is there any way to have a quick fix without using the master?
I am concerned about the potential changes I have to do in my code for going from the version I am using (1.4.0) and the master.

mvelebit · 2020-03-04T14:45:29Z

@eduardo4jesus You can explicitly add headers as stated above, something alike:

opener = urllib.request.URLopener()
opener.addheader('User-Agent', some_user_agent)
opener.retrieve(
    url, fpath,
    reporthook=gen_bar_updater()
)

(line 81 and onwards in vision/torchvision/datasets/utils.py). Seems to be a quick workaround that works.

See pytorch/vision#1938 Signed-off-by: Nicolas V Castet <[email protected]>

nvcastet · 2020-03-04T15:56:40Z

@eduardo4jesus You could patch your model script at the top using:

from six.moves import urllib
opener = urllib.request.build_opener()
opener.addheaders = [('User-agent', 'Mozilla/5.0')]
urllib.request.install_opener(opener)

It will use that user agent for the entire script assuming the opener does not get overwritten somewhere else.

nvcastet · 2020-03-04T18:44:28Z

To make it work for python 2 as well:

import urllib
try:
    # For python 2
    class AppURLopener(urllib.FancyURLopener):
        version = "Mozilla/5.0"

    urllib._urlopener = AppURLopener()
except AttributeError:
    # For python 3
    opener = urllib.request.build_opener()
    opener.addheaders = [('User-agent', 'Mozilla/5.0')]
    urllib.request.install_opener(opener)

joergsimon · 2020-03-04T19:18:38Z

so for python 3 I now use the following snipplet:

from torchvision import datasets
import torchvision.transforms as transforms
import urllib

num_workers = 0
batch_size = 20
basepath = 'some/base/path'
transform = transforms.ToTensor()

def set_header_for(url, filename):
    opener = urllib.request.URLopener()
    opener.addheader('User-Agent', 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_9_3) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/35.0.1916.47 Safari/537.36')
    opener.retrieve(
    url, f'{basepath}/{filename}')

set_header_for('http://yann.lecun.com/exdb/mnist/train-images-idx3-ubyte.gz', 'train-images-idx3-ubyte.gz')
set_header_for('http://yann.lecun.com/exdb/mnist/train-labels-idx1-ubyte.gz', 'train-labels-idx1-ubyte.gz')
set_header_for('http://yann.lecun.com/exdb/mnist/t10k-images-idx3-ubyte.gz', 't10k-images-idx3-ubyte.gz')
set_header_for('http://yann.lecun.com/exdb/mnist/t10k-labels-idx1-ubyte.gz', 't10k-labels-idx1-ubyte.gz')
train_data = datasets.MNIST(root='data', train=True,
                                   download=True, transform=transform)
test_data = datasets.MNIST(root='data', train=False,
                                  download=False, transform=transform)

You would need to modify the basepath variable of course

fatihbeyhan · 2020-03-04T20:13:12Z

I've just got the same problem. Waiting for the answer without changing codes... (ROOKIE ALERT)

knamdar · 2020-03-05T01:17:42Z

I've just got the same problem. Waiting for the answer without changing codes... (ROOKIE ALERT)

Clone this to your working dir:
https://github.com/knamdar/data

joergsimon · 2020-03-05T12:34:57Z

The problem ist that Yann LeCun’s side changed hoster if I got it right, and this one checks if the HTTP headers are set. I currently work around with the following code: from torchvision import datasets import torchvision.transforms as transforms import urllib num_workers = 0 batch_size = 20 basepath = 'some/base/path' transform = transforms.ToTensor() def set_header_for(url, filename): opener = urllib.request.URLopener() opener.addheader('User-Agent', 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_9_3) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/35.0.1916.47 Safari/537.36') opener.retrieve( url, f'{basepath}/{filename}') set_header_for('http://yann.lecun.com/exdb/mnist/train-images-idx3-ubyte.gz', 'train-images-idx3-ubyte.gz') set_header_for('http://yann.lecun.com/exdb/mnist/train-labels-idx1-ubyte.gz', 'train-labels-idx1-ubyte.gz') set_header_for('http://yann.lecun.com/exdb/mnist/t10k-images-idx3-ubyte.gz', 't10k-images-idx3-ubyte.gz') set_header_for('http://yann.lecun.com/exdb/mnist/t10k-labels-idx1-ubyte.gz', 't10k-labels-idx1-ubyte.gz') train_data = datasets.MNIST(root='data', train=True, download=True, transform=transform) test_data = datasets.MNIST(root='data', train=False, download=False, transform=transform) You need to change base path of course

…

On 05.03.2020, at 05:26, Nikita Makarin ***@***.***> wrote: I've the same issue when I'm trying to get datasets: import torch import torchvision from torchvision import transforms, datasets train = datasets.MNIST("", train=True, download=True, transform=transforms.Compose([transforms.ToTensor()])) test = datasets.MNIST("", train=False, download=True, transform=transforms.Compose([transforms.ToTensor()])) — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#1938?email_source=notifications&email_token=AAN2AFNSOADTTTO6F3JRBLDRF4SZFA5CNFSM4LBCIY62YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEN3VCJQ#issuecomment-595022118>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAN2AFI4ZQEJJ2HEPJCBHP3RF4SZFANCNFSM4LBCIY6Q>.

eduardo4jesus · 2020-03-05T16:28:53Z

from six.moves import urllib
opener = urllib.request.build_opener()
opener.addheaders = [('User-agent', 'Mozilla/5.0')]
urllib.request.install_opener(opener)

@nvcastet, Thank you so much for the clarification. At that point I misunderstood that I would have to go into Torchvision library and change one of its internal files, which would not ben a smooth move on Colab/Kaggle.

vision/torchvision/datasets/utils.py

fmassa · 2020-03-05T16:51:21Z

This should have been fixed now, there is no need to update torchvision.

All should be working as before, without any change on the user side.

This was fixed on the server hosting the original dataset (thanks @soumith !).

As such, I'm closing this issue but let us know if you still face this issue.

junpuf · 2021-03-11T19:37:16Z

Did anyone got urllib.error.HTTPError: HTTP Error 503: Service Unavailable last night?

ChengguiSun · 2021-03-11T21:34:57Z

copy this snippet at the top of your notebook, run it, and then just load your datasets as usual...
from six.moves import urllib
opener = urllib.request.build_opener()
opener.addheaders = [('User-agent', 'Mozilla/5.0')]
urllib.request.install_opener(opener)

Didn't work for me on Colab today

Didn't work for me either

BernardoOlisan · 2021-03-12T03:33:34Z

Did anyone got urllib.error.HTTPError: HTTP Error 503: Service Unavailable last night?

Yep yesterday it was working for me but now is not, I dont know what happend, is there a solution for this?

alisterburt · 2021-03-12T08:35:58Z

@BernardoOlisan @ChengguiSun the solution from @mlelarge works great, the following worked for me in a notebook

!wget www.di.ens.fr/~lelarge/MNIST.tar.gz
!tar -zxvf MNIST.tar.gz

from torchvision.datasets import MNIST
from torchvision import transforms

mnist_train = MNIST('./', download=False,
                    transform=transforms.Compose([
                        transforms.ToTensor(),
                    ]), train=True)

HugoSchmutz · 2021-03-12T08:54:33Z

Thank you ! @mlelarge and @alisterburt

BernardoOlisan · 2021-03-12T14:01:47Z

!wget www.di.ens.fr/~lelarge/MNIST.tar.gz
!tar -zxvf MNIST.tar.gz

nope still got the exact same error

alisterburt · 2021-03-12T14:05:23Z

@BernardoOlisan do you get the same error when trying to download from that link in a browser? Just tried again locally and all worked fine...

scaomath · 2021-03-12T19:18:58Z

!wget www.di.ens.fr/~lelarge/MNIST.tar.gz
!tar -zxvf MNIST.tar.gz
nope still got the exact same error

You have to have a folder called MNIST/raw/ in the folder with your dataset script or notebook

torchvision.datasets.MNIST(root='./', download=False,
                    transform=transforms.Compose([
                        transforms.ToTensor(),
                    ]), train=True)

If you have your code

torchvision.datasets.MNIST(root='./data', download=False,
                    transform=transforms.Compose([
                        transforms.ToTensor(),
                    ]), train=True)

then the folder has to be structured as

├── data
│   ├── MNIST
│         ├── processed
│         └── raw
└── Python script

andrewssobral · 2021-03-12T19:41:08Z

The solution proposed by @mlelarge works fine for me, but in my case I have:

# For new versions of TorchVision
wget -O MNIST.tar.gz https://activeeon-public.s3.eu-west-2.amazonaws.com/datasets/MNIST.new.tar.gz

# For old versions of TorchVision (<= 1.0.1*)
wget -O MNIST.tar.gz https://activeeon-public.s3.eu-west-2.amazonaws.com/datasets/MNIST.old.tar.gz

tar -zxvf MNIST.tar.gz

then, I do:

...
train_loader = torch.utils.data.DataLoader(
  datasets.MNIST('./', train=True, download=False,
                 transform=transforms.Compose([
                     transforms.ToTensor(),
                     transforms.Normalize((0.1307,), (0.3081,))
                 ])),
  batch_size=args.batch_size, shuffle=True, **kwargs)
...

and everything works well.

fmassa · 2021-03-12T21:11:24Z

This should be fixed (again) in the next torchvision nightly, and the fix will be present in the next minor release of torchvision, which should be out soon.

See #3544 for more details

andrewssobral · 2021-03-12T22:51:05Z

Perfect, thank you @fmassa !

ChengguiSun · 2021-03-17T06:05:04Z

@alisterburt, thank you. I used that solution too. Sometimes it works, sometimes it doesn't. It works for me tonight. It's just very slow to download the dataset.

@BernardoOlisan @ChengguiSun the solution from @mlelarge works great, the following worked for me in a notebook

!wget www.di.ens.fr/~lelarge/MNIST.tar.gz
!tar -zxvf MNIST.tar.gz

from torchvision.datasets import MNIST
from torchvision import transforms

mnist_train = MNIST('./', download=False,
                    transform=transforms.Compose([
                        transforms.ToTensor(),
                    ]), train=True)

mohamed-180 · 2021-03-21T09:47:05Z

try this

from torchvision import datasets
datasets.MNIST(".", download=True)

scaomath · 2021-03-25T13:32:05Z

@alisterburt, thank you. I used that solution too. Sometimes it works, sometimes it doesn't. It works for me tonight. It's just very slow to download the dataset.

Because www.di.ens.fr/~lelarge is apparently @mlelarge 's personal website's storage on a university's web host which probably cannot handle the bandwidth for all amateur data scientists relying on torch.dataset. We should download the original from LeCun's site manually by ourselves, upload to a personal cloud storage that supports http, then use wget.

EricPengShuai · 2021-08-04T11:10:19Z

@BernardoOlisan @ChengguiSun the solution from @mlelarge works great, the following worked for me in a notebook

!wget www.di.ens.fr/~lelarge/MNIST.tar.gz
!tar -zxvf MNIST.tar.gz

from torchvision.datasets import MNIST
from torchvision import transforms

mnist_train = MNIST('./', download=False,
                    transform=transforms.Compose([
                        transforms.ToTensor(),
                    ]), train=True)

@alisterburt I encountered the following problems after trying

SYSTEM_WGETRC = c:/progra~1/wget/etc/wgetrc
syswgetrc = C:\Download\GnuWin32-wget/etc/wgetrc
--2021-08-04 19:11:37--  http://www.di.ens.fr/~lelarge/MNIST.tar.gz
正在解析主机 www.di.ens.fr... 129.199.99.14
Connecting to www.di.ens.fr|129.199.99.14|:80... 已连接。
已发出 HTTP 请求，正在等待回应... 302 Found
位置：https://www.di.ens.fr/~lelarge/MNIST.tar.gz [跟随至新的 URL]
--2021-08-04 19:11:37--  https://www.di.ens.fr/~lelarge/MNIST.tar.gz
Connecting to www.di.ens.fr|129.199.99.14|:443... 已连接。
无法建立 SSL 连接。
tar: Error opening archive: Failed to open 'MNIST.tar.gz'
---------------------------------------------------------------------------
RuntimeError                              Traceback (most recent call last)
<ipython-input-2-48b8be9a8697> in <module>
      8                     transform=transforms.Compose([
      9                         transforms.ToTensor(),
---> 10                     ]), train=True)

c:\users\peng\anaconda3\envs\pytorch_1.8\lib\site-packages\torchvision\datasets\mnist.py in __init__(self, root, train, transform, target_transform, download)
     80
     81         if not self._check_exists():
---> 82             raise RuntimeError('Dataset not found.' +
     83                                ' You can use download=True to download it')
     84

RuntimeError: Dataset not found. You can use download=True to download it

rodrsilv-stark · 2021-08-04T11:30:39Z

Have you tried to decompress the file first?

EricPengShuai · 2021-08-04T11:36:53Z

Have you tried to decompress the file first?

No, I just try it in the ipython(anaconda/pytorch 1.8) in my windows. But I try it in my macbook successfully. I don't know why.

SiddhantSadangi · 2024-05-14T15:27:37Z

@Borda - seems like this issue has resurfaced

fancellu · 2024-05-14T17:27:52Z

I try

dataset = torchvision.datasets.MNIST(root="mnist/", train=True, download=True, transform=torchvision.transforms.ToTensor())

and I get 403s

Downloading http://yann.lecun.com/exdb/mnist/train-images-idx3-ubyte.gz
Failed to download (trying next):
HTTP Error 403: Forbidden```

dandv · 2024-05-23T20:30:03Z

Still getting this error with torchvision==0.16.2.

trainset = torchvision.datasets.MNIST(
    "./downloads/mnist",
    download=True,
    train=True,
)

Downloading http://yann.lecun.com/exdb/mnist/train-images-idx3-ubyte.gz
Failed to download (trying next):
HTTP Error 403: Forbidden

Downloading http://yann.lecun.com/exdb/mnist/train-labels-idx1-ubyte.gz
Failed to download (trying next):
HTTP Error 403: Forbidden

mahtin · 2024-07-01T06:53:05Z

This is still happening intermittently when reading the files from http://yann.lecun.com/exdb/mnist/ however, this simple fix solves the intermittent nature of this issue.

< datasets_url = 'http://yann.lecun.com/exdb/mnist/'
> datasets_url = 'https://storage.googleapis.com/cvdf-datasets/mnist/'

The same files are present there and in fact many places, for example: https://github.com/golbin/TensorFlow-MNIST/raw/master/mnist/data/ plus other places.

I think it's a sad case of the original site being either overrun or broken.

vanjy-eng · 2024-08-09T03:25:02Z

Still seeing this issue on Mac OS.
torch==2.0.1
torchvision==0.15.2

None of the above mentioned methods worked.

Code:

train_dataset = datasets.MNIST('./data', train=True, download=True, transform=transform)
test_dataset = datasets.MNIST('./data', train=False, transform=transform)

Fixed using this stackoverflow issue:
SSL issue

rudrOwO · 2024-10-05T08:24:01Z

Yes, finding it impossible to bypass the 403 error. I've tried using aforementioned custom headers, but no luck

fedhere · 2024-10-06T15:34:32Z

I can bypass the 403 error on colab, but hit a HTTPError 302, which as I understand means I am getting redirected but the redirect fails, which may be dues still to cookies not being set correctly

import ssl
ssl._create_default_https_context = ssl._create_unverified_context

from torchvision import datasets
import torchvision.transforms as transforms
import urllib

num_workers = 0
batch_size = 20
basepath = '.'
transform = transforms.ToTensor()

def set_header_for(url, filename):
    opener = urllib.request.URLopener()
    opener.addheader('User-Agent', 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_9_3) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/35.0.1916.47 Safari/537.36')
    opener.retrieve(
    url, f'{basepath}/{filename}')

set_header_for('http://yann.lecun.com/exdb/mnist/train-images-idx3-ubyte.gz', 'train-images-idx3-ubyte.gz')
set_header_for('http://yann.lecun.com/exdb/mnist/train-labels-idx1-ubyte.gz', 'train-labels-idx1-ubyte.gz')
set_header_for('http://yann.lecun.com/exdb/mnist/t10k-images-idx3-ubyte.gz', 't10k-images-idx3-ubyte.gz')
set_header_for('http://yann.lecun.com/exdb/mnist/t10k-labels-idx1-ubyte.gz', 't10k-labels-idx1-ubyte.gz')
train_data = datasets.MNIST(root='data', train=True,
                                   download=True, transform=transform)
test_data = datasets.MNIST(root='data', train=False,
                                  download=False, transform=transform)
                                  
 --------------------------------------------------------
 
 <ipython-input-11-7b9c22e51c89>:19: DeprecationWarning: URLopener style of invoking requests is deprecated. Use newer urlopen functions/methods
  set_header_for('http://yann.lecun.com/exdb/mnist/train-images-idx3-ubyte.gz', 'train-images-idx3-ubyte.gz')
---------------------------------------------------------------------------
HTTPError                                 Traceback (most recent call last)
[<ipython-input-11-7b9c22e51c89>](https://localhost:8080/#) in <cell line: 19>()
     17     url, f'{basepath}/{filename}')
     18 
---> 19 set_header_for('http://yann.lecun.com/exdb/mnist/train-images-idx3-ubyte.gz', 'train-images-idx3-ubyte.gz')
     20 set_header_for('http://yann.lecun.com/exdb/mnist/train-labels-idx1-ubyte.gz', 'train-labels-idx1-ubyte.gz')
     21 set_header_for('http://yann.lecun.com/exdb/mnist/t10k-images-idx3-ubyte.gz', 't10k-images-idx3-ubyte.gz')

6 frames
[<ipython-input-11-7b9c22e51c89>](https://localhost:8080/#) in set_header_for(url, filename)
     14     opener = urllib.request.URLopener()
     15     opener.addheader('User-Agent', 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_9_3) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/35.0.1916.47 Safari/537.36')
---> 16     opener.retrieve(
     17     url, f'{basepath}/{filename}')
     18 

[/usr/lib/python3.10/urllib/request.py](https://localhost:8080/#) in retrieve(self, url, filename, reporthook, data)
   1819             except OSError:
   1820                 pass
-> 1821         fp = self.open(url, data)
   1822         try:
   1823             headers = fp.info()

[/usr/lib/python3.10/urllib/request.py](https://localhost:8080/#) in open(self, fullurl, data)
   1785         try:
   1786             if data is None:
-> 1787                 return getattr(self, name)(url)
   1788             else:
   1789                 return getattr(self, name)(url, data)

[/usr/lib/python3.10/urllib/request.py](https://localhost:8080/#) in open_http(self, url, data)
   1963     def open_http(self, url, data=None):
   1964         """Use HTTP protocol."""
-> 1965         return self._open_generic_http(http.client.HTTPConnection, url, data)
   1966 
   1967     def http_error(self, url, fp, errcode, errmsg, headers, data=None):

[/usr/lib/python3.10/urllib/request.py](https://localhost:8080/#) in _open_generic_http(self, connection_factory, url, data)
   1957                               response.status)
   1958         else:
-> 1959             return self.http_error(
   1960                 url, response.fp,
   1961                 response.status, response.reason, response.msg, data)

[/usr/lib/python3.10/urllib/request.py](https://localhost:8080/#) in http_error(self, url, fp, errcode, errmsg, headers, data)
   1979                 result = method(url, fp, errcode, errmsg, headers, data)
   1980             if result: return result
-> 1981         return self.http_error_default(url, fp, errcode, errmsg, headers)
   1982 
   1983     def http_error_default(self, url, fp, errcode, errmsg, headers):

[/usr/lib/python3.10/urllib/request.py](https://localhost:8080/#) in http_error_default(self, url, fp, errcode, errmsg, headers)
   1984         """Default error handler: close the connection and raise OSError."""
   1985         fp.close()
-> 1986         raise HTTPError(url, errcode, errmsg, headers, None)
   1987 
   1988     if _have_ssl:

HTTPError: HTTP Error 302: Found

This was referenced Mar 4, 2020

hparams as dict Lightning-AI/pytorch-lightning#1029

Merged

create own MNIST dataset Lightning-AI/pytorch-lightning#942

Closed

fmassa added bug help wanted module: datasets labels Mar 4, 2020

Borda mentioned this issue Mar 4, 2020

Add MNIST dataset & drop torchvision dep. from tests Lightning-AI/pytorch-lightning#986

Merged

5 tasks

ptrblck added a commit to ptrblck/vision that referenced this issue Mar 4, 2020

fix for pytorch#1938

8f8c751

ptrblck mentioned this issue Mar 4, 2020

Add header for MNIST download due to cloudflare browser check #1939

Closed

nvcastet added a commit to nvcastet/horovod that referenced this issue Mar 4, 2020

Patch pytorch_mnist.py until torchvision issue gets resolved

ce85225

See pytorch/vision#1938 Signed-off-by: Nicolas V Castet <[email protected]>

nvcastet mentioned this issue Mar 4, 2020

Patch pytorch_mnist.py until torchvision issue gets resolved horovod/horovod#1768

Merged

Borda mentioned this issue Mar 4, 2020

fix MNIST download Lightning-AI/pytorch-lightning#1044

Merged

vandanavk mentioned this issue Mar 4, 2020

Update smdebug to 0.7.0 aws/sagemaker-pytorch-training-toolkit#172

Merged

seemethere mentioned this issue Mar 4, 2020

datasets: Fallback to our own mirrors for mnist #1940

Closed

zhangguanheng66 mentioned this issue Mar 5, 2020

mnist example is failing pytorch/examples#725

Closed

fmassa closed this as completed Mar 5, 2020

tfjgeorge mentioned this issue Mar 16, 2021

adds dot product to PVector tfjgeorge/nngeometry#16

Merged

alar0330 mentioned this issue Mar 21, 2021

[SageMaker Experiments] fix MNIST download issue with torchvision aws/amazon-sagemaker-examples#2091

Merged

mitchjablonski mentioned this issue Apr 9, 2021

Intro To Pytorch Download Issues udacity/deep-learning-v2-pytorch#316

Closed

akihironitta mentioned this issue May 6, 2021

Downloading MNIST fails due to HTTP Error 503 Lightning-Universe/lightning-bolts#632

Closed

arbfay mentioned this issue Oct 13, 2021

Pytorch Tutorial 5 - MNIST classification - Jupyter and Markdown Adamage/tutorials#91

Merged

PolyachenkoYA mentioned this issue Jul 25, 2024

Install fixes NachiStern/MNIST-Contrasive-Local-Learning-Networks#1

Merged

Downloading MNIST dataset with torchvision gives HTTP Error 403 #1938

Downloading MNIST dataset with torchvision gives HTTP Error 403 #1938

Comments

Borda commented Mar 4, 2020

🐛 Bug

To Reproduce

Environment

Additional context

fmassa commented Mar 4, 2020

fmassa commented Mar 4, 2020 • edited Loading

soumith commented Mar 4, 2020

Borda commented Mar 4, 2020

fmassa commented Mar 4, 2020

eduardo4jesus commented Mar 4, 2020 • edited Loading

mvelebit commented Mar 4, 2020 • edited Loading

nvcastet commented Mar 4, 2020 • edited Loading

nvcastet commented Mar 4, 2020

joergsimon commented Mar 4, 2020 • edited Loading

fatihbeyhan commented Mar 4, 2020

knamdar commented Mar 5, 2020

joergsimon commented Mar 5, 2020 via email

eduardo4jesus commented Mar 5, 2020 • edited Loading

fmassa commented Mar 5, 2020 • edited Loading

junpuf commented Mar 11, 2021

ChengguiSun commented Mar 11, 2021

BernardoOlisan commented Mar 12, 2021

alisterburt commented Mar 12, 2021 • edited Loading

HugoSchmutz commented Mar 12, 2021

BernardoOlisan commented Mar 12, 2021

alisterburt commented Mar 12, 2021

scaomath commented Mar 12, 2021 • edited Loading

andrewssobral commented Mar 12, 2021 • edited Loading

fmassa commented Mar 12, 2021

andrewssobral commented Mar 12, 2021

ChengguiSun commented Mar 17, 2021

mohamed-180 commented Mar 21, 2021

scaomath commented Mar 25, 2021

EricPengShuai commented Aug 4, 2021 • edited Loading

rodrsilv-stark commented Aug 4, 2021

EricPengShuai commented Aug 4, 2021

SiddhantSadangi commented May 14, 2024 • edited Loading

fancellu commented May 14, 2024 • edited Loading

dandv commented May 23, 2024 • edited Loading

mahtin commented Jul 1, 2024

vanjy-eng commented Aug 9, 2024

rudrOwO commented Oct 5, 2024

fedhere commented Oct 6, 2024

fmassa commented Mar 4, 2020 •

edited

Loading

eduardo4jesus commented Mar 4, 2020 •

edited

Loading

mvelebit commented Mar 4, 2020 •

edited

Loading

nvcastet commented Mar 4, 2020 •

edited

Loading

joergsimon commented Mar 4, 2020 •

edited

Loading

eduardo4jesus commented Mar 5, 2020 •

edited

Loading

fmassa commented Mar 5, 2020 •

edited

Loading

alisterburt commented Mar 12, 2021 •

edited

Loading

scaomath commented Mar 12, 2021 •

edited

Loading

andrewssobral commented Mar 12, 2021 •

edited

Loading

EricPengShuai commented Aug 4, 2021 •

edited

Loading

SiddhantSadangi commented May 14, 2024 •

edited

Loading

fancellu commented May 14, 2024 •

edited

Loading

dandv commented May 23, 2024 •

edited

Loading