Skip to content

armgit / dspy

Public Imported from GitHub
0

DSPy: The framework for programming—not prompting—language models

Python

This PR adds [chromadb](https://docs.trychroma.com/getting-started) to the RM.

Insop Songbf14c3c616a4
diff --git a/README.md b/README.md
index a38564c99..e710c99fa 100644
--- a/README.md
+++ b/README.md
@@ -70,6 +70,12 @@ For the optional Qdrant retrieval integration, include the `qdrant` extra:
pip install dspy-ai[qdrant]
```
+For the optional [chromadb](https://github.com/chroma-core/chroma) retrieval integration, include the `chromadb` extra:
+
+```
+pip install dspy-ai[chromadb]
+```
+
## 2) Syntax: You're in charge of the workflow—it's free-form Python code!
**DSPy** hides tedious prompt engineering, but it cleanly exposes the important decisions you need to make: **[1]** what's your system design going to look like? **[2]** what are the important constraints on the behavior of your program?
diff --git a/dspy/retrieve/chromadb_rm.py b/dspy/retrieve/chromadb_rm.py
index b6f515a9b..82ab0023f 100644
--- a/dspy/retrieve/chromadb_rm.py
+++ b/dspy/retrieve/chromadb_rm.py
@@ -11,8 +11,6 @@ from dsp.utils import dotdict
try:
import chromadb
from chromadb.config import Settings
-
- # from chromadb.utils import embedding_functions
except ImportError:
chromadb = None
@@ -27,7 +25,7 @@ class ChromadbRM(dspy.Retrieve):
A retrieval module that uses chromadb to return the top passages for a given query.
Assumes that the chromadb index has been created and populated with the following metadata:
- - text: The text of the passage
+ - documents: The text of the passage
Args:
collection_name (str): chromadb collection name
@@ -41,16 +39,18 @@ class ChromadbRM(dspy.Retrieve):
dspy.Prediction: An object containing the retrieved passages.
Examples:
- Below is a code snippet that shows how to use this as the default retriver:
+ Below is a code snippet that shows how to use this as the default retriever:
```python
llm = dspy.OpenAI(model="gpt-3.5-turbo")
- retriever_model = ChromadbRM('collection_name', 'db_path', openai.api_key, 'azure')
+ retriever_model = ChromadbRM('collection_name', 'db_path')
dspy.settings.configure(lm=llm, rm=retriever_model)
+ # to test the retriever with "my query"
+ retriever_model("my query")
```
Below is a code snippet that shows how to use this in the forward() function of a module
```python
- self.retrieve = ChromadbRM(k=num_passages)
+ self.retrieve = ChromadbRM('collection_name', 'db_path', k=num_passages)
```
"""
@@ -66,15 +66,11 @@ class ChromadbRM(dspy.Retrieve):
openai_api_version: Optional[str] = None,
k: int = 7,
):
- # self._openai_embed_model = embedding_functions.OpenAIEmbeddingFunction(
- # model_name=openai_embed_model
- # )
-
self._openai_embed_model = openai_embed_model
self._init_chromadb(collection_name, persist_directory)
- # If not provided, defaults to env vars OPENAI_API_KEY and OPENAI_ORGANIZATION
+ # If not provided, defaults to env vars
if openai_api_key:
openai.api_key = openai_api_key
if openai_api_type:
@@ -171,8 +167,6 @@ class ChromadbRM(dspy.Retrieve):
query_embeddings=embeddings, n_results=k
)
- # 'metadatas': [[{'page': 30, 'source': "..."}, {...}]]
passages = [dotdict({"long_text": x}) for x in results["documents"][0]]
- print(passages)
return passages