Andrew Luo

Also encountered this issue when porting the code, I believe the min of the bounding box for a room can be used as an origin. This can be found as...

Just a heads up, code for extracting a dense (not 14 x 14 patch) feature map [was shared here](https://github.com/ShirAmir/dino-vit-features/blob/main/extractor.py) from a pretrained ViT. There are two caveats to using their...

I should be able to help on the high quality video data aspect (along with transcripts), although the source of captions is a more difficult problem.

This is still an issue @danthe3rd More serious as conda currently default ships with python 3.11.7, but xformers conda only supports up to 3.11.0.

@yil8 You need to set scene.set_flags((1

The corruption fix was for a bug introduced in zstd 1.5.0 (see zstd bug 3517), it seems that zfs uses zstd 1.4.5 with upstream patches ported by @ryao zstd versioning...

@behlendorf I was curious if you knew how btrfs dealt with this. Btrfs uses kernel zstd, which have been updated multiple times in the last few years. How do they...

So latent vector (torch.nn.Embedding) optimization has had a few issues for a few years now. Apparently it is because if momentum/decay gets applied to all elements in an embedding object,...

Thanks! I guess my confusion is what the radius parameter in `nn = pynanoflann.KDTree(n_neighbors=5, metric='L1', radius=100)` does. Edit: Also curious what the canonical behavior for `_, indices = nn.radius_neighbors(query, radius=distance)`...

For the radius query, there seems to be a minor bug. The python interface is returning square rooted distance, but the radius query takes in a squared query.