Credit: Samsung
The breakfasts I was able to identify cluster into three major regions:
，更多细节参见同城约会
This month, OpenAI announced their Codex app and my coworkers were asking questions. So I downloaded it, and as a test case for the GPT-5.2-Codex (high) model, I asked it to reimplement the UMAP algorithm in Rust. UMAP is a dimensionality reduction technique that can take in a high-dimensional matrix of data and simultaneously cluster and visualize data in lower dimensions. However, it is a very computationally-intensive algorithm and the only tool that can do it quickly is NVIDIA’s cuML which requires CUDA dependency hell. If I can create a UMAP package in Rust that’s superfast with minimal dependencies, that is an massive productivity gain for the type of work I do and can enable fun applications if fast enough.
Notice how the highlighted region shrinks at each step. The algorithm never examines points outside the narrowing window. In a balanced tree with nnn points, this takes about log4(n)\log_4(n)log4(n) steps. For a million points, that's roughly 10 steps instead of a million comparisons.