DSPy can't retrieve Passage with Text Embeddings in ChromaDB

[ananduaji]

I am working on a RAG application using DSPy and ChromaDB for pdf files.

At first I fetched the text from the pdf and add it to the Chromadb as chunks. Also added the embeddings of the chunks.

like this

def store_document_in_chromadb(text):
    chunks = chunk_document(text)
    ids = [f'chunk_{i}' for i in range(len(chunks))]
    embeddings = [get_embedding(chunk).tolist() for chunk in chunks]

    collection.add(ids=ids, documents=chunks, embeddings=embeddings)

And I try to retrieve the relevant chunks like this,

retriever_model = ChromadbRM("contracts_collection", 'db/', k=2)
dspy.settings.configure(lm=llama2_model, rm=retriever_model)

class GenerateAnswer(dspy.Signature): 
    """Answer the question based on the context given."""
    context = dspy.InputField(desc="may contain relevant context")
    question = dspy.InputField()
    answer = dspy.OutputField(desc="often between 5 to 10 words")

class RAG(dspy.Module): 
    def __init__(self, num_passages=2):
        super().__init__()
        self.retrieve = dspy.Retrieve(k=num_passages)
        self.generate_answer = dspy.ChainOfThought(GenerateAnswer)
    
    def forward(self, question):
        context = self.retrieve(question).passages
        prediction = self.generate_answer(context=context, question=question)
        return dspy.Prediction(context=context, answer=prediction.answer)

with dspy.context(lm=llama2_model, rm=retriever_model):
  module = RAG()
  response = module("What is the Total Spend")
  print(response)

When I a running this, getting this error

InvalidDimensionException: Embedding dimension 384 does not match collection dimensionality 768

but when I remove the embedding from the ChromaDB, its retrieving the relevant chunks correctly.

Does anyone know why is this not getting while using embeddings?