2 ms·
I have a question: what hardware did you use and how long did you need to generate the embeddings ?
by simlevesque 10mo ago
I have a question: what hardware did you use and how long did you need to generate the embeddings ?
- afiodorov 10mo agoDaily updates I do on my m4 mac air: takes about 5 minutes to process roughly 10k fresh comments. Historic backfill was done on an Nvidia GPU rented on vast.ai for a few dollars. If I recall correctly took about an hour or so. It’s mentioned in the README.md on GitHub.