vanvalenlab / deepcell-label

Cloud-based data annotation tools for biological images
https://label.deepcell.org
Other
73 stars 14 forks source link

Unable to load images while using DeepCell-label locally #522

Closed ArozHada closed 1 year ago

ArozHada commented 1 year ago

Describe the bug Hello, I can get DeepCell-label welcome page running locally but I am unable to upload any images. I cannot also open any example images locally. I get an error on my backend terminal saying " X.ome.tiff not found in zip. Loading all tiffs in zip. y.ome.tiff not found in zip. Loading all tiffs in zip. Warning: JSON file divisions.json not found. Warning: JSON file lineage.json not found. Warning: JSON file cellTypes.json not found. Warning: JSON file cells.json not found. Warning: JSON file embeddings.json not found. [2023-08-21 16:48:33,064] ERROR in blueprints: Encountered ValueError: axes do not match shape Traceback (most recent call last): File "/Users/anaconda3/envs/deepcell_label_env/lib/python3.9/site-packages/flask/app.py", line 1516, in full_dispatch_request rv = self.dispatch_request()

botocore.exceptions.ClientError: An error occurred (AuthorizationHeaderMalformed) when calling the PutObject operation: The authorization header is malformed; a non-empty Access Key (AKID) must be provided in the credential."

I have also tried uploading a zip folder with tiffs but I encounter the same error. Is there any specific format for the upload of the images or did I miss something in my local installation?

Furthermore, can I upload a segmentation mask along with the corresponding image?

To Reproduce Steps to reproduce the behavior:

  1. Run frontend and backend as suggested in the documentation for local use.
  2. Upload any image when using the application locally
  3. See error

Expected behaviour Opening of the annotation studio

Desktop (please complete the following information):

rossbar commented 1 year ago

In order to run locally, you will still need to provide a bucket that data will be uploaded to. URLs and credentials are specified in the .env file.

ArozHada commented 1 year ago

Is there any way to run Deepcell Label without using an aws bucket and using local storage?

ykevu commented 1 year ago

The backend codebase would have to be changed to allow for that, as it assumes that a bucket is provided. As far as I know an older version of DeepCell Label supported local usage without an AWS bucket, so another solution would be to roll back your clone to such a version (at the obvious cost of some bugs existing and none of the new features for cell types)

ArozHada commented 1 year ago

Ok, Thank you for the quick response. :)

ykevu commented 1 year ago

No problem--to be a little more specific, I started working on DCL just after the 0.4.0 release, which had already moved to AWS-only for local, so definitely don't try anything from then on