Jobs are stuck, and cannot be aborted

my jobs are stuck, and i cant abort them

for example:

i click the “Abort build” button, and nothing happens


now the jobs are running ok, but i am getting an error message that indicates that the dataset is not on the server:

/workspace/venv/lib/python3.7/site-packages/torchvision/datasets/ RuntimeWarning: The use of the download flag is deprecated, since the dataset is no longer publicly accessible.
  warnings.warn(msg, RuntimeWarning)
Benchmarking TResNet
Changing number of GPUs to 1 for server 

Benchmarking on ImageNet...
Traceback (most recent call last):
  File "/workspace/venv/lib/python3.7/site-packages/torchbench/image_classification/", line 185, in benchmark
  File "/workspace/venv/lib/python3.7/site-packages/torchvision/datasets/", line 45, in __init__
    raise RuntimeError(msg)
RuntimeError: The dataset is no longer publicly accessible. You need to download the archives externally and place them in the root directory.

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/workspace/repository/", line 57, in <module>
  File "/workspace/venv/lib/python3.7/site-packages/sotabenchapi/core/", line 85, in regular_evaluation
    func(*args, **kwargs)
  File "/workspace/venv/lib/python3.7/site-packages/torchbench/image_classification/", line 193, in benchmark
  File "/workspace/venv/lib/python3.7/site-packages/torchvision/datasets/", line 54, in __init__
  File "/workspace/venv/lib/python3.7/site-packages/torchvision/datasets/", line 69, in parse_archives
  File "/workspace/venv/lib/python3.7/site-packages/torchvision/datasets/", line 148, in parse_devkit_archive
    _verify_archive(root, file, md5)
  File "/workspace/venv/lib/python3.7/site-packages/torchvision/datasets/", line 102, in _verify_archive
    raise RuntimeError(msg.format(file, root))
RuntimeError: The archive ILSVRC2012_devkit_t12.tar.gz is not present in the root directory or is corrupted. You need to download it externally and place it in ./.data/vision/imagenet.

Hi! Sorry about this. There was a temporary problem with the AWS instances getting stuck, but I think everything is working now.

About the dataset - yes, this is a recent change in torchvision, the ImageNet dataset cannot be automatically downloaded any more. You’ll need to download it yourself with your own script, or you can also upload it directly to your user cache folder using the sb command from the sotabench-api python library (also here: The sb command has command-line documentation that should guide you through the process.

thanks, solved the issue