[lxml] Optimizing lxml for Handling Large XML Files: Tips and Experiences

I've been using lxml to process large XML files recently, and I'm looking for ways to optimize performance. Specifically, I'm trying to filter specific nodes from large datasets and manage memory usage more effectively. I'd appreciate any tips or best practices from your experiences. Are there any techniques you use to enhance performance, or potential pitfalls I should watch out for? Thanks in advance for your insights!