dspy
dspy copied to clipboard
stanfordnlp
DSPy: The framework for programming—not prompting—language models
Make the parameters `return_all_scores` and `return_outputs` of the Evaluate class more reasonable. 1. If I specify `return_outputs=False` when calling `__call__`, this value is not actually passed in, as `return_outputs` will...
I am trying to build an Agent using Function Calling and try to implement it with DSPy, but I haven't found the instructions for using DSPy with function calling.
Hello, Not that it's a critical error, but I'm getting a conflict with the Pydantic version required by dspy-ai and other critical packages necessary for my application as follows: dspy-ai...
This PR introduces several helper methods for the `Module` and `Predict` classes in order to allow for convenient ways to: 1. Check if two modules (e.g. a student and a...
There are many llm api calls during pipeline optimization. Does DSPY support asynchronous calls to LLMs yet? When trainset size up to hundreds level, the current sequential execution is too...
I'm trying to compile a zeroshot chain using BootstrapFewShot and langchain. I've followed [this](https://github.com/stanfordnlp/dspy/blob/main/examples/tweets/compiling_langchain.ipynb) and [this](https://python.langchain.com/v0.2/docs/integrations/providers/dspy/), with the only difference being that I'm using my own retriever. ``` optimizer =...
Hello DSPy Team, I want to request the addition of support for OpenAI's structured output in your library. OpenAI recently introduced structured outputs in their API, which seems to guarantee...
## 📝 Changes Description This MR/PR contains the following changes: - Add a seperate module in `dsp` for Anthropic via VertexAI as described here: https://cloud.google.com/vertex-ai/generative-ai/docs/partner-models/use-claude#use_the_anthropic_sdk This is just a rough...
Metadata
Owner
Metadata
DSPy: The framework for programming—not prompting—language models