Daniel Crane
Daniel Crane
Same here. @cbalint13 Doesn't the 2013 code require MATLAB in order to be able to run it, though? :(
It's just a visualisation error, it just takes "person" since that's the first class in the default coco dataset. There are multiple possible ways you could fix it, to name...
> @danielcrane Many thanks. > I replaced the CLASSES part in [https://github.com/WXinlong/SOLO/blob/master/mmdet/datasets/coco.py](https://github.com/WXinlong/SOLO/blob/master/mmdet/datasets/coco.py?rgh-link-date=2021-08-25T04%3A36%3A22Z) with CLASSES=('bird'), but it doesn't work and I got the "person" label again :(( The classes are actually...
Another place to try could be in `test_ins_vis.py` on the following line: ```python class_names = get_classes("coco") ``` you could either try to change it to your new dataset name, or...
@azade-a Happy that I could help, sorry it took us so long to get there!
By the way, I personally had issues with the model not learning well when training with one class, so if you also encounter the same issue then my recommendation would...
Same for me too, did you manage to fix this issue?
@lauwg I managed to "fix" it by adding a second dummy class, I couldn't get it to work with just one class.
No idea I'm afraid, only ever tried on GPU!
@mathmax12 Where is the `AssertionError` coming from? It looks like an `mmdet`-related error - do you already have another version of `mmdet` installed that could be interfering with the one...