mirror of
https://github.com/hwchase17/langchain
synced 2024-11-08 07:10:35 +00:00
72bfc1d3db
This PR migrates the existing MongoDBAtlasVectorSearch abstraction from
the `langchain_community` section to the partners package section of the
codebase.
- [x] Run the partner package script as advised in the partner-packages
documentation.
- [x] Add Unit Tests
- [x] Migrate Integration Tests
- [x] Refactor `MongoDBAtlasVectorStore` (autogenerated) to
`MongoDBAtlasVectorSearch`
- [x] ~Remove~ deprecate the old `langchain_community` VectorStore
references.
## Additional Callouts
- Implemented the `delete` method
- Included any missing async function implementations
- `amax_marginal_relevance_search_by_vector`
- `adelete`
- Added new Unit Tests that test for functionality of
`MongoDBVectorSearch` methods
- Removed [`del
res[self._embedding_key]`](e0c81e1cb0/libs/community/langchain_community/vectorstores/mongodb_atlas.py (L218))
in `_similarity_search_with_score` function as it would make the
`maximal_marginal_relevance` function fail otherwise. The `Document`
needs to store the embedding key in metadata to work.
Checklist:
- [x] PR title: Please title your PR "package: description", where
"package" is whichever of langchain, community, core, experimental, etc.
is being modified. Use "docs: ..." for purely docs changes, "templates:
..." for template changes, "infra: ..." for CI changes.
- Example: "community: add foobar LLM"
- [x] PR message
- [x] Pass lint and test: Run `make format`, `make lint` and `make test`
from the root of the package(s) you've modified to check that you're
passing lint and testing. See contribution guidelines for more
information on how to write/run tests, lint, etc:
https://python.langchain.com/docs/contributing/
- [x] Add tests and docs: If you're adding a new integration, please
include
1. Existing tests supplied in docs/docs do not change. Updated
docstrings for new functions like `delete`
2. an example notebook showing its use. It lives in
`docs/docs/integrations` directory. (This already exists)
If no one reviews your PR within a few days, please @-mention one of
baskaryan, efriis, eyurtsev, hwchase17.
---------
Co-authored-by: Steven Silvester <steven.silvester@ieee.org>
Co-authored-by: Erick Friis <erick@langchain.dev>
88 lines
3.1 KiB
Python
88 lines
3.1 KiB
Python
"""
|
|
Tools for the Maximal Marginal Relevance (MMR) reranking.
|
|
Duplicated from langchain_community to avoid cross-dependencies.
|
|
|
|
Functions "maximal_marginal_relevance" and "cosine_similarity"
|
|
are duplicated in this utility respectively from modules:
|
|
- "libs/community/langchain_community/vectorstores/utils.py"
|
|
- "libs/community/langchain_community/utils/math.py"
|
|
"""
|
|
|
|
import logging
|
|
from typing import List, Union
|
|
|
|
import numpy as np
|
|
|
|
logger = logging.getLogger(__name__)
|
|
|
|
Matrix = Union[List[List[float]], List[np.ndarray], np.ndarray]
|
|
|
|
|
|
def cosine_similarity(X: Matrix, Y: Matrix) -> np.ndarray:
|
|
"""Row-wise cosine similarity between two equal-width matrices."""
|
|
if len(X) == 0 or len(Y) == 0:
|
|
return np.array([])
|
|
|
|
X = np.array(X)
|
|
Y = np.array(Y)
|
|
if X.shape[1] != Y.shape[1]:
|
|
raise ValueError(
|
|
f"Number of columns in X and Y must be the same. X has shape {X.shape} "
|
|
f"and Y has shape {Y.shape}."
|
|
)
|
|
try:
|
|
import simsimd as simd # type: ignore
|
|
|
|
X = np.array(X, dtype=np.float32)
|
|
Y = np.array(Y, dtype=np.float32)
|
|
Z = 1 - simd.cdist(X, Y, metric="cosine")
|
|
if isinstance(Z, float):
|
|
return np.array([Z])
|
|
return Z
|
|
except ImportError:
|
|
logger.info(
|
|
"Unable to import simsimd, defaulting to NumPy implementation. If you want "
|
|
"to use simsimd please install with `pip install simsimd`."
|
|
)
|
|
X_norm = np.linalg.norm(X, axis=1)
|
|
Y_norm = np.linalg.norm(Y, axis=1)
|
|
# Ignore divide by zero errors run time warnings as those are handled below.
|
|
with np.errstate(divide="ignore", invalid="ignore"):
|
|
similarity = np.dot(X, Y.T) / np.outer(X_norm, Y_norm)
|
|
similarity[np.isnan(similarity) | np.isinf(similarity)] = 0.0
|
|
return similarity
|
|
|
|
|
|
def maximal_marginal_relevance(
|
|
query_embedding: np.ndarray,
|
|
embedding_list: list,
|
|
lambda_mult: float = 0.5,
|
|
k: int = 4,
|
|
) -> List[int]:
|
|
"""Calculate maximal marginal relevance."""
|
|
if min(k, len(embedding_list)) <= 0:
|
|
return []
|
|
if query_embedding.ndim == 1:
|
|
query_embedding = np.expand_dims(query_embedding, axis=0)
|
|
similarity_to_query = cosine_similarity(query_embedding, embedding_list)[0]
|
|
most_similar = int(np.argmax(similarity_to_query))
|
|
idxs = [most_similar]
|
|
selected = np.array([embedding_list[most_similar]])
|
|
while len(idxs) < min(k, len(embedding_list)):
|
|
best_score = -np.inf
|
|
idx_to_add = -1
|
|
similarity_to_selected = cosine_similarity(embedding_list, selected)
|
|
for i, query_score in enumerate(similarity_to_query):
|
|
if i in idxs:
|
|
continue
|
|
redundant_score = max(similarity_to_selected[i])
|
|
equation_score = (
|
|
lambda_mult * query_score - (1 - lambda_mult) * redundant_score
|
|
)
|
|
if equation_score > best_score:
|
|
best_score = equation_score
|
|
idx_to_add = i
|
|
idxs.append(idx_to_add)
|
|
selected = np.append(selected, [embedding_list[idx_to_add]], axis=0)
|
|
return idxs
|