Lue Fan

It's hard to say what is going wrong. You could list the detailed procedure you followed, including gata generation, the code you use, and modifications you made if any. I...

Sorry for late reply. How is it going now? 1. I don't know, it's hard to say. Check the IO or timing to find the bottleneck. 2. Increase learning rate...

FSD++ is our new work, which does not simply concatenate 6-frame point clouds. We are going to release it one month later.

Below is the validation results of plain FSD uses 6 past frames, trained with 6 epochs. I have not tried the better 12-epoch schedule yet. Hope it helps. ``` OBJECT_TYPE_TYPE_VEHICLE_LEVEL_1:...

torch_scatter is not included in TorchEx. It is from https://github.com/rusty1s/pytorch_scatter. Have you successfully deployed spconv, which I think is the most important thing.

I will not recommend you deploy a sparse conv based algorithm if you are not an experienced engineer... There are many things you need to deal with before deploying it,...

Yes, I believe all the inputs should be in fixed sizes.

Why not try smaller channels or depth?

For training data, there is no need to conduct forward or backward extensions. Otherwise, the bin file may reach the size limit.

1. In the tracking config, you should set `max_time_since_update` to 0 for training data generation. Otherwise, you will get an output with huge size. 2. Yes, just replace `waymo_infos_val.pkl` with...