Show HN: MonkeyPatch – Cheap, fast and predictable LLM functions in Python

Re:Show HN: MonkeyPatch – Cheap, fast and predictable LLM fu... by p10jkle on Wednesday November 15, 2023 @11:25AM (#38278465)

Nice! If I were to write a test for invariant aspects of the function (eg, it produces valid json), will the system guarantee that those invariants are fulfilled? I suppose naively you could just do this by calling over and over and 'telling off' the model if it didn't get it right

[ Parent | on HN ]

Re:Show HN: MonkeyPatch – Cheap, fast and predictable LLM fu... by JackHopkins on Wednesday November 15, 2023 @12:06PM (#38279139)

The type constraints are indeed enforced but not by the tests but by the type-hints you give to the patched functions. The constraints and enforced structure are followed, there is also a repair feedback loop in place if the original LLM output is invalid for the types you've declared. Tests are more to align the model how to act for different inputs. Hope this makes it more clear!

[ Parent | on HN ]