Vision: Bad zipfile error while downloading CelebA dataset

Created on 5 Oct 2020  路  9Comments  路  Source: pytorch/vision

馃悰 Bug

I am working with CelebA dataset with download = True. I am getting an error
BadZipFile: File is not a zip file

To Reproduce

import torchvision.datasets as dset
data = dset.celeba.CelebA(root='.', split='train', target_type='attr', transform=None, download=True)

Environment

I am working on Google Colab
```
PyTorch version: 1.6.0+cu101
Is debug build: False
CUDA used to build PyTorch: 10.1
ROCM used to build PyTorch: N/A

OS: Ubuntu 18.04.5 LTS (x86_64)
GCC version: (Ubuntu 7.5.0-3ubuntu1~18.04) 7.5.0
Clang version: 6.0.0-1ubuntu2 (tags/RELEASE_600/final)
CMake version: version 3.12.0

Python version: 3.6 (64-bit runtime)
Is CUDA available: True
CUDA runtime version: 10.1.243
GPU models and configuration: GPU 0: Tesla K80
Nvidia driver version: 418.67
cuDNN version: /usr/lib/x86_64-linux-gnu/libcudnn.so.7.6.5
HIP runtime version: N/A
MIOpen runtime version: N/A

Versions of relevant libraries:
[pip3] numpy==1.18.5
[pip3] torch==1.6.0+cu101
[pip3] torchsummary==1.5.1
[pip3] torchtext==0.3.1
[pip3] torchvision==0.7.0+cu101
[conda] Could not collect```

cc @pmeier

duplicate datasets

Most helpful comment

@square-1111 can you try doing in a new interpreter

import tqdm

and report back? You probably don't have tqdm installed, and the PyTorch fallback doesn't implement close

@pmeier we should probably fix this somehow, either by removing the need of a close in our offending function or by adding a close to the fallback implementation in PyTorch

All 9 comments

This is a duplicate of https://github.com/pytorch/vision/issues/2387 and https://github.com/pytorch/vision/issues/2317, and a "fix" is done in https://github.com/pytorch/vision/pull/2321

The issue is that CelebA is stored in Google Drive, and it has exceeded the download quota for the day.
Try again in a different moment and it should work

Thank You for the leads.

AttributeError: 'tqdm' object has no attribute 'close' -- I am getting this error on local machine. Is this error associated with exceeding download quota or something else?

@square-1111 could you please provide a minimal code snippet and list of installed packages to reproduce the issue. Thanks

This error is different -- can you check if you have tqdm installed? If not, might be that the helper implementation of tqdm that we provide doesn't have the close method

@vfdev-5 I am trying to run
import torchvision.datasets as dset img_path ='.' data = dset.celeba.CelebA(root=img_path, split='train', target_type='attr', transform=None, download=True)

and getting this error AttributeError: 'tqdm' object has no attribute 'close'

@square-1111 can you try doing in a new interpreter

import tqdm

and report back? You probably don't have tqdm installed, and the PyTorch fallback doesn't implement close

@pmeier we should probably fix this somehow, either by removing the need of a close in our offending function or by adding a close to the fallback implementation in PyTorch

I'll look into it.

pytorch/pytorch#46040 will fix the tqdm mock.

Was this page helpful?
0 / 5 - 0 ratings

Related issues

varagrawal picture varagrawal  路  3Comments

Wadaboa picture Wadaboa  路  3Comments

feiyangsuo picture feiyangsuo  路  3Comments

Linardos picture Linardos  路  4Comments

300LiterPropofol picture 300LiterPropofol  路  3Comments