Ash Vardanian

Gotcha, @davvard . Can you please patch it into a more readable form with 4 columns, so that its easier to reason about it. --- Is it a good idea...

Gotcha! Full compatibility is great! And how do they generate the IDs, @davvard ?

@davvard That sounds ugly... as we would either need to store a counter for every pair... or retrieve the list of all the neighbors every time. Which IDs will NetworkX...

@criver, thank you for the request! It is definitely one of our top priorities right now. Will keep you posted. [Feel free to join our Discord for updates](https://discord.gg/4mxGrenbNt).

Efficient Python implementation would require finishing #41

The construction part may be just a direct call to `ukv_docs_import` and `ukv_graph_import_t` from #154. After the construction, sampling (#153) should be performed to randomly selected items to be queried...

I also find those to be exciting datasets. I have used both of them when comparing the quality of cross-modal "semantic joins" with uni-modal embeddings [in one of my recent...

@maumueller sounds good! Let's split the work into 2 halves. First, merge USearch, then - new datasets. As for USearch, I am considering a few options for the `config.yml` and...

Hey, @erikbern and @maumueller! Thank you for following up! As of right now, I'm in the active development phase of [USearch v3](https://github.com/unum-cloud/usearch/pull/277) and [UCall v1](https://github.com/unum-cloud/ucall), hoping to finish the first...

@maumueller, I was also thinking about the `big-ann-benchmarks`, but this year they've shrunk the dataset size from 1 Billion down to tens of Millions... We have a new result coming...