Ilias MAOUDJ

Hello, I know this post is quite old but I think the [tips](https://github.com/facebookresearch/votenet/blob/master/doc/tips.md) given are not complete. More than modifying the input to 3D rotation, I think we must modify...

Downgrading to Pytorch 1.1 and using python 3.6 worked for me.

> @alar0330 Yeah this was the problem in extract_split.m file. I'm facing issues in extract_rgbd_data_v2.m file In extract_rgbd_data_v2.m, I changed the parfor loop to a for loop since Octave (without...

I think that if you want to retrain the model to detect cup you should modify (lower) the radius of the set abstraction in models/backbone_module.py

Check your load_mask function and if you do the right "if statement test" `if image_info["source"] != "YOUR_SOURCE_NAME": return super(self.__class__, self).load_mask(image_id)`