Closed Nelsordo closed 1 year ago
Most likely it was introduced in this PR: https://github.com/heartexlabs/label-studio-converter/pull/157/files
@p0p4k could you please help with this issue?
@Nelsordo Could you switch to this branch and check how it works please? https://github.com/heartexlabs/label-studio-converter/pull/195
I downloaded the branch and run python setup.py install
and then label-studio-converter import yolo -i C:\Users\Nelson\Downloads\oilspilldetection.v2i.yolov5pytorch\aaa -o output.json
but keep sending the same .json file
Maybe is a step where I am doing it wrong, could you please give me a step by step of how to import pre-annotated Dataset in Yolov5 format to Label-Studio using Local Storage? I would appreciate it a lot.
Here are more details. I'm using Win11 and an enviroment in Anaconda3 to use LS. In the documentation says that we have to change this variable LABEL_STUDIO_LOCAL_FILES_DOCUMENT_ROOT=C:\Users\Nelson
@Nelsordo Are you sure you switched to the PR branch?
cd label-studio-converter
git fetch origin fix/yolo-import-image-root
git checkout fix/yolo-import-image-root
pip install -U -e . # almost the same as python setup.py install
label-studio-converter import yolo -i .....
Following your instructions the converter has worked diferrent, it now throw me this .json and it keep sending an error
Did you add source in the project settings? Add the source for your image files and do not click sync.
@p0p4k try again, I made some change:
git pull origin fix/yolo-import-image-root
pip install -U -e .
label-studio-converter import yolo -i .....
Please try the latest master branch, now it should be fixed.
Follow up on this, slightly different but same issue, how would you make use of the label studio converter with an S3 storage system? I am having the same URL error and can't seem to use label-studio converter properly on my system
Hi, I've an issue when I import pre-annotated data with yolo format to LS. How you can see it send me an error and the image is not annotated. I have tried many ways but none have worked. This are my steps.
label-studio-converter import yolo -i C:\Users\Nelson\Downloads\oilspilldetection.v2i.yolov5pytorch\aaa -o output.json
I've tried to import the dataset (converted) by local-host and it produces the same error
The .json file is the following: And the paths of the files are: "C:\Users\Nelson\Downloads\oilspilldetection.v2i.yolov5pytorch\aaa\images\USR_shadow_2437_fake_png.rf.8d21fc53f30644d6e0952c348b6805c1.jpg" "C:\Users\Nelson\Downloads\oilspilldetection.v2i.yolov5pytorch\aaa\images\output.json"