Image Discrepancy between Model Training and Dataset

Hi everyone,

I’ve come across a discrepancy in my dataset that I can’t quite explain.

In the main dataset, I have 660 images. However, when I create a version, it shows 659 total images. The missing image’s name is “ptcntmh791 - 2025-07-09_transformed_20250805_122011.JPG”. The confusing part is that when I sum the images across the splits (train + valid + test), I still get 660 images.

So:

  • Dataset: 660 images

  • Version total shown: 659 images

  • Train + Valid + Test: 660 images

Any ideas on what might be causing this or how to debug it would be really helpful.

Thanks!

  • Project Type: Instance Segmentation
  • Operating System & Browser: Windows & Google Chrome
  • Project Universe Link or Workspace/Project ID: fixed-quadrats-pt-2-qjt6o
  • Do you grant Roboflow Support permission to access your Workspace for troubleshooting? (Yes/No): Yes