Chris Choy

I used TitanRTX with 24G VRAM. To put the image on the GPU, you could adjust other settings such as the image size, or number of hardest negative mining samples.

Not sure what you are doing inside the make sparse tensor function, but if you want to force the output to use a specific key, feed the key as an...

The only problematic part is the openmp, which MinGW could solve. However, I haven't tested it.

I see. It seems like a good feature.

The first error seems like pytorch did not use ninja automatically for compilation. Please use the latest master, which forces ninja for compilation, instead of the release version. ``` git...

It seems like you are not using the latest master. Copy paste this command to get it ``` git clone https://github.com/NVIDIA/MinkowskiEngine.git ```

It's unlikely to think that there's memory leak when over 10k iterations, the memory fluctuates from 21.2-21.7. When there's a memory leak, the memory consumption CONSTANTLY increases. Can you share...

I've never seen a memory leak that doesn't leak until 10k iterations but starts to leak after 10k iterations. This looks more like OS level memory management and garbage collection...

Voxel size determines the resolution of the space. Let's say that we have a 100m x 100m x 25m LIDAR scan. If we select voxel size to be | voxel...

The reason you got the same number is because you used `TensorField`. This is a wrapper for continuous point cloud. You can use .sparse() to convert `TensorField` to `SparseTensor` which...