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
Hello, thanks for the great dataset and amazing experiment documentation in the paper.
While working on this dataset I noticed that the training annotations contain 65 duplicate images with different image IDs in livecell_coco_train.json listed below as name and id 1 and id 2.
I checked a few of the duplicates to see if they have different annotations but it seems that they don't.
Furthermore, I'm not sure if this is intended but the validation set is not mutually exclusive with the training set. There are 30 common images between them:
Hello, thanks for the great dataset and amazing experiment documentation in the paper.
While working on this dataset I noticed that the training annotations contain 65 duplicate images with different image IDs in
livecell_coco_train.json
listed below as name and id 1 and id 2.The validation file
livecell_coco_val.json
has 1 duplicate.The test file
livecell_coco_test.json
has 52 duplicates.I checked a few of the duplicates to see if they have different annotations but it seems that they don't.
Furthermore, I'm not sure if this is intended but the validation set is not mutually exclusive with the training set. There are 30 common images between them:
The text was updated successfully, but these errors were encountered: