You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
It looks like the json files are very large and after the "image_path" keyword, they seem to contain strange strings.
Could it be that these files also contain the images as well?
The text was updated successfully, but these errors were encountered:
We may refactor PolygonDataset to use that data instead of jpegs at some point, or release new data with that issue fixed, but for now I've just pushed 89d0406 which adds scripts/prune_imageData.py and some associated notes in the docs/README.
I think keeping images as jpegs and the labels as jsons would make it easier to version the dataset. Simple diffs on the jsons are possible and makes it overall more lightweight.
I have a question regarding the assets.zip.
It looks like the json files are very large and after the "image_path" keyword, they seem to contain strange strings.
Could it be that these files also contain the images as well?
The text was updated successfully, but these errors were encountered: