Removing magic from openai functions #2
Merged
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Thanks to the feedback from brad0577 in our Discord channel:
I agree, I got carried away when implementing OpenAI functions, but this is not the philosophy for LiteChain. So indeed I am removing this functionality from LiteChain, and
functions
will now receive a list of dicts with the schema as the official API does.Unfortunately this will be a breaking change, hopefully for quite few people at this point. To cover up a bit, I'm adding an example in the docs using the openai_function_call schema to at least cover the schema extraction part.
Going forward, if we want to have the chain-calling-another-chain directly functionality, it would be better to be something else composing on top of
OpenAIChatChain
, and not ingrained into itThanks brad!
[1] Link to the updated docs on OpenAI function calling
[2] Link to the example using openai_function_call to extract the schema