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
I would like to use gbnet to generate a scene graph for an image that is not in VG. To do so, it looks I need to use mode = sgdet. I am modifying one of the .ipynb files in eval_sdet/. However, it looks like in the detector files, pred_entries is filled in by caches results during training. Any help on how to just generate a scene graph without any dependent files from training.
The text was updated successfully, but these errors were encountered:
I would like to use gbnet to generate a scene graph for an image that is not in VG. To do so, it looks I need to use mode = sgdet. I am modifying one of the .ipynb files in eval_sdet/. However, it looks like in the detector files, pred_entries is filled in by caches results during training. Any help on how to just generate a scene graph without any dependent files from training.
Hi!
I would like to use gbnet to generate a scene graph for an image that is not in VG. To do so, it looks I need to use mode = sgdet. I am modifying one of the .ipynb files in eval_sdet/. However, it looks like in the detector files, pred_entries is filled in by caches results during training. Any help on how to just generate a scene graph without any dependent files from training.
The text was updated successfully, but these errors were encountered: