pytorch / pytorch/rl

[BUG] AtariDQNExperienceReplay shows AccessDeniedException

Open
#3,113 1 comment 0 reactions 1 assignee View on GitHub

@vmoens is already working on this.

Since Aug 6, 2025.

bug
Dominant language
Python
Stars
3.6k
Forks
487
Avg merge
1d 1h
Merged PRs (30d)
207

Description

Describe the bug

Trying to download a supposedly available dataset, as described in the API documentation, leads to a "AccessDeniedException" from gsutils.

This happens as a logged-in and anonymous user in the gcloud cli.

To Reproduce

Executing the following lines, shown in the linked API reference page below, results in the mentioned exception:

>>> from torchrl.data.datasets import AtariDQNExperienceReplay
>>> dataset = AtariDQNExperienceReplay("Pong/5", batch_size=128)

https://docs.pytorch.org/rl/0.9/reference/generated/torchrl.data.datasets.AtariDQNExperienceReplay.html#torchrl.data.datasets.AtariDQNExperienceReplay

These is the line in the code that can not be executed successfully:

command = f"gsutil -m ls -R gs://atari-replay-datasets/dqn/{self.dataset_id}/replay_logs"
output = subprocess.run(
    command, shell=True, capture_output=True
)  # , stdout=subprocess.DEVNULL, stderr=subprocess.DEVNULL)

The full error (as logged-in user via glcoud cli) is:

AccessDeniedException: 403 does not have storage.objects.list access to the Google Cloud Storage bucket. Permission 'storage.objects.list' denied on resource

Expected behavior

A successful download

Screenshots

If applicable, add screenshots to help explain your problem.

System info

Describe the characteristic of your environment:

import torchrl, numpy, sys
print(torchrl.__version__, numpy.__version__, sys.version, sys.platform)

0.9.2 2.3.2 3.13.2 | packaged by Anaconda, Inc. | (main, Feb 6 2025, 18:56:02) [GCC 11.2.0] linux

Additional context

Add any other context about the problem here.

Reason and Possible fixes

Seems like the folder on gcloud is not publicly available.
Not sure if it ever was, but the code and the documentation makes it seem like it should be.

Checklist

  • I have checked that there is no similar issue in the repo (required)
  • I have read the documentation (required)
  • I have provided a minimal working example to reproduce the bug (required)

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Assessment

This issue has not been assessed yet.

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.