Rob Meng

> what kind of improvement do you expect from it? so It is mainly save memory save/load, iiuc? > It is just that this 5% seems a bit small to...

> @jn2clark could you set spawn method > > ``` > from multiprocessing import set_start_method > set_start_method("spawn") > ``` > > Before running Pytorch loader? to add some color here....

There is no way to create index without data. The index is `IVF_PQ`, which requires data to train a k-means clustering.

What is your use case, i.e. how many vectors, how good of a recall number do you want to see? how frequently do you make updates? We can try to...

Hi, could you provide the version of `lance` you are trying to use. Also, steps to reproduce would be helpful.

Hi @reymondzzzz @rakshith-ravi I believe the GHA runner you are using are both intel macs, which we don't support. Could you try running on `macos-14`? https://docs.github.com/en/actions/using-github-hosted-runners/about-github-hosted-runners/about-github-hosted-runners#supported-runners-and-hardware-resources

potentially interesting solution: https://github.com/rust-lang/api-guidelines/discussions/95 (via feature flag)

https://github.com/EmbarkStudios/rust-ecosystem/blob/main/lints.toml might be a good list to start with

Hi, What's the platform/OS you are running on?

What's the distro? Context: we only release manylinux 2_24 for aarch64 linux.