-
Notifications
You must be signed in to change notification settings - Fork 47
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Cut Lab 7 - Semantic Search runtime to 15 minutes #46
Comments
One option would be to have this lab be the very last one. We fire off the notebook and then wrap. |
An alternative would be to find another example to showcase the vector search, ideally with the Form 13 data set the labs focus on. |
The current lab seems to want the embeddings to be pre computed. That's the Julia Child thing we were avoiding in these labs, instead walking through how everything was done so it would be real. |
The semantic search lab is going to use the same Form 13 data set we use elsewhere. We'll compute the embedding in the GDS lab as before. But now we'll store it using the vector index and explore that functionality. Finally we'll export it to a CSV for later consumption by AutoML. |
We're dropping the GDS lab. Instead we ought to find a way to cut runtime for this lab down to 15 minutes. We could prebake the embeddings but it'd be better if we could walk attendees through everything. |
Ezhil cut the lab down and that works now in a real account. But the free trial has a quota that we're hitting. Trying to figure, out a workaround. |
Used a throttle to work around the quota issue but now a single cell takes 15 minutes to run. Ezhil looking at ways to cut it down. |
It has a note: "# Primary loop. This could take 30 minutes to an hour." That's not going to work in the labs. Need to come up with an alternative solution.
The text was updated successfully, but these errors were encountered: