Stuck on an issue?

Lightrun Answers was designed to reduce the constant googling that comes with debugging 3rd party libraries. It collects links to all the places you might be looking at while hunting down a tough bug.

And, if you’re still stuck at the end, we’re happy to hop on a call to see how we can help out.

Cost estimation of using GPT-3 user simulator

See original GitHub issue

Definition of Done:

Compute the cost incurred to generate conv-set-A and conv-set-B (Including the ones that were not labelled as either of these tags but were still generated in the effort to get to these two sets of conversations).
Statistics on number of conversations, length of conversations, number of conversations not selected to be tagged under conv-set-A / conv-set-B.
Create an enhancement issue for weizenbot to report the cost incurred in generating each dialogue. https://github.com/RasaHQ/weizenbot/issues/34

Issue Analytics

State:
Created 2 years ago
Comments:9 (9 by maintainers)

Top GitHub Comments

1reaction

koernerfeliciacommented, Oct 12, 2021

Yep, that’s what I was thinking too, though I can just parse the tracker rather than creating another rep, and I should use the 4 tokens per 3 words estimate like listed here to avoid under-estimating

0reactions

koernerfeliciacommented, Oct 18, 2021

Yep, sorry, I thought I had linked it in the definition of done:

https://github.com/RasaHQ/weizenbot/issues/34