geoleva

There are several problems with this. First, for Vis Genome model, the labels (150 total) are abstract (LABEL_0 to LABEL_149). I checked, and OI model's config labels are also abstract....

when executing __load_pretrained_model_, here is what I see for **missing_keys** (which are the modules in initialized Deformable DETR - modules loaded from checkpoint dict states). Essentially, these are modules that...

> It looks like DETR weights were used rather than EGTR weights. Would you like to try again using this [link](https://drive.google.com/file/d/18phcRxbrEI7HqIuM2OLAPuwAF5k3pUC2/view)? @jinbae - this is the same file, it does...

@icecap360 and @jinbae: I think I somewhat figured out the steps: (1) The Deformable DETR model is initialized via call: model = DetrForSceneGraphGeneration.from_pretrained( args.architecture, config=config, ignore_mismatched_sizes=True ) At this point,...

@jinbae - found 150 labels used in VG-150 dataset. Still, would be nice to have the config.json fixed with these. I am attaching the updated config I did. So, only...

@jinbae - : 1. There is no folder "**dataset**" in the EGTR commit, and hence those JSON files aren't accessible. Is it possible to commit them? I think I figured...

The Visual Genome project distribution does **not** seem to have this file - at least I could not find it. It has the **object.json** file, but it's the annotations of...