Spaces:
Runtime error
Runtime error
File size: 15,209 Bytes
58d33f0 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 |
"""Wrapper around FAISS vector database."""
from __future__ import annotations
import pickle
import uuid
from pathlib import Path
from typing import Any, Callable, Dict, Iterable, List, Optional, Tuple
import numpy as np
from langchain.docstore.base import AddableMixin, Docstore
from langchain.docstore.document import Document
from langchain.docstore.in_memory import InMemoryDocstore
from langchain.embeddings.base import Embeddings
from langchain.vectorstores.base import VectorStore
from langchain.vectorstores.utils import maximal_marginal_relevance
def dependable_faiss_import() -> Any:
"""Import faiss if available, otherwise raise error."""
try:
import faiss
except ImportError:
raise ValueError(
"Could not import faiss python package. "
"Please install it with `pip install faiss` "
"or `pip install faiss-cpu` (depending on Python version)."
)
return faiss
class FAISS(VectorStore):
"""Wrapper around FAISS vector database.
To use, you should have the ``faiss`` python package installed.
Example:
.. code-block:: python
from langchain import FAISS
faiss = FAISS(embedding_function, index, docstore, index_to_docstore_id)
"""
def __init__(
self,
embedding_function: Callable,
index: Any,
docstore: Docstore,
index_to_docstore_id: Dict[int, str],
):
"""Initialize with necessary components."""
self.embedding_function = embedding_function
self.index = index
self.docstore = docstore
self.index_to_docstore_id = index_to_docstore_id
def __add(
self,
texts: Iterable[str],
embeddings: Iterable[List[float]],
metadatas: Optional[List[dict]] = None,
**kwargs: Any,
) -> List[str]:
if not isinstance(self.docstore, AddableMixin):
raise ValueError(
"If trying to add texts, the underlying docstore should support "
f"adding items, which {self.docstore} does not"
)
documents = []
for i, text in enumerate(texts):
metadata = metadatas[i] if metadatas else {}
documents.append(Document(page_content=text, metadata=metadata))
# Add to the index, the index_to_id mapping, and the docstore.
starting_len = len(self.index_to_docstore_id)
self.index.add(np.array(embeddings, dtype=np.float32))
# Get list of index, id, and docs.
full_info = [
(starting_len + i, str(uuid.uuid4()), doc)
for i, doc in enumerate(documents)
]
# Add information to docstore and index.
self.docstore.add({_id: doc for _, _id, doc in full_info})
index_to_id = {index: _id for index, _id, _ in full_info}
self.index_to_docstore_id.update(index_to_id)
return [_id for _, _id, _ in full_info]
def add_texts(
self,
texts: Iterable[str],
metadatas: Optional[List[dict]] = None,
**kwargs: Any,
) -> List[str]:
"""Run more texts through the embeddings and add to the vectorstore.
Args:
texts: Iterable of strings to add to the vectorstore.
metadatas: Optional list of metadatas associated with the texts.
Returns:
List of ids from adding the texts into the vectorstore.
"""
if not isinstance(self.docstore, AddableMixin):
raise ValueError(
"If trying to add texts, the underlying docstore should support "
f"adding items, which {self.docstore} does not"
)
# Embed and create the documents.
embeddings = [self.embedding_function(text) for text in texts]
return self.__add(texts, embeddings, metadatas, **kwargs)
def add_embeddings(
self,
text_embeddings: Iterable[Tuple[str, List[float]]],
metadatas: Optional[List[dict]] = None,
**kwargs: Any,
) -> List[str]:
"""Run more texts through the embeddings and add to the vectorstore.
Args:
text_embeddings: Iterable pairs of string and embedding to
add to the vectorstore.
metadatas: Optional list of metadatas associated with the texts.
Returns:
List of ids from adding the texts into the vectorstore.
"""
if not isinstance(self.docstore, AddableMixin):
raise ValueError(
"If trying to add texts, the underlying docstore should support "
f"adding items, which {self.docstore} does not"
)
# Embed and create the documents.
texts = [te[0] for te in text_embeddings]
embeddings = [te[1] for te in text_embeddings]
return self.__add(texts, embeddings, metadatas, **kwargs)
def similarity_search_with_score_by_vector(
self, embedding: List[float], k: int = 4
) -> List[Tuple[Document, float]]:
"""Return docs most similar to query.
Args:
query: Text to look up documents similar to.
k: Number of Documents to return. Defaults to 4.
Returns:
List of Documents most similar to the query and score for each
"""
scores, indices = self.index.search(np.array([embedding], dtype=np.float32), k)
docs = []
for j, i in enumerate(indices[0]):
if i == -1:
# This happens when not enough docs are returned.
continue
_id = self.index_to_docstore_id[i]
doc = self.docstore.search(_id)
if not isinstance(doc, Document):
raise ValueError(f"Could not find document for id {_id}, got {doc}")
docs.append((doc, scores[0][j]))
return docs
def similarity_search_with_score(
self, query: str, k: int = 4
) -> List[Tuple[Document, float]]:
"""Return docs most similar to query.
Args:
query: Text to look up documents similar to.
k: Number of Documents to return. Defaults to 4.
Returns:
List of Documents most similar to the query and score for each
"""
embedding = self.embedding_function(query)
docs = self.similarity_search_with_score_by_vector(embedding, k)
return docs
def similarity_search_by_vector(
self, embedding: List[float], k: int = 4, **kwargs: Any
) -> List[Document]:
"""Return docs most similar to embedding vector.
Args:
embedding: Embedding to look up documents similar to.
k: Number of Documents to return. Defaults to 4.
Returns:
List of Documents most similar to the embedding.
"""
docs_and_scores = self.similarity_search_with_score_by_vector(embedding, k)
return [doc for doc, _ in docs_and_scores]
def similarity_search(
self, query: str, k: int = 4, **kwargs: Any
) -> List[Document]:
"""Return docs most similar to query.
Args:
query: Text to look up documents similar to.
k: Number of Documents to return. Defaults to 4.
Returns:
List of Documents most similar to the query.
"""
docs_and_scores = self.similarity_search_with_score(query, k)
return [doc for doc, _ in docs_and_scores]
def max_marginal_relevance_search_by_vector(
self, embedding: List[float], k: int = 4, fetch_k: int = 20
) -> List[Document]:
"""Return docs selected using the maximal marginal relevance.
Maximal marginal relevance optimizes for similarity to query AND diversity
among selected documents.
Args:
embedding: Embedding to look up documents similar to.
k: Number of Documents to return. Defaults to 4.
fetch_k: Number of Documents to fetch to pass to MMR algorithm.
Returns:
List of Documents selected by maximal marginal relevance.
"""
_, indices = self.index.search(np.array([embedding], dtype=np.float32), fetch_k)
# -1 happens when not enough docs are returned.
embeddings = [self.index.reconstruct(int(i)) for i in indices[0] if i != -1]
mmr_selected = maximal_marginal_relevance(
np.array([embedding], dtype=np.float32), embeddings, k=k
)
selected_indices = [indices[0][i] for i in mmr_selected]
docs = []
for i in selected_indices:
if i == -1:
# This happens when not enough docs are returned.
continue
_id = self.index_to_docstore_id[i]
doc = self.docstore.search(_id)
if not isinstance(doc, Document):
raise ValueError(f"Could not find document for id {_id}, got {doc}")
docs.append(doc)
return docs
def max_marginal_relevance_search(
self, query: str, k: int = 4, fetch_k: int = 20
) -> List[Document]:
"""Return docs selected using the maximal marginal relevance.
Maximal marginal relevance optimizes for similarity to query AND diversity
among selected documents.
Args:
query: Text to look up documents similar to.
k: Number of Documents to return. Defaults to 4.
fetch_k: Number of Documents to fetch to pass to MMR algorithm.
Returns:
List of Documents selected by maximal marginal relevance.
"""
embedding = self.embedding_function(query)
docs = self.max_marginal_relevance_search_by_vector(embedding, k, fetch_k)
return docs
def merge_from(self, target: FAISS) -> None:
"""Merge another FAISS object with the current one.
Add the target FAISS to the current one.
Args:
target: FAISS object you wish to merge into the current one
Returns:
None.
"""
if not isinstance(self.docstore, AddableMixin):
raise ValueError("Cannot merge with this type of docstore")
# Numerical index for target docs are incremental on existing ones
starting_len = len(self.index_to_docstore_id)
# Merge two IndexFlatL2
self.index.merge_from(target.index)
# Create new id for docs from target FAISS object
full_info = []
for i in target.index_to_docstore_id:
doc = target.docstore.search(target.index_to_docstore_id[i])
if not isinstance(doc, Document):
raise ValueError("Document should be returned")
full_info.append((starting_len + i, str(uuid.uuid4()), doc))
# Add information to docstore and index_to_docstore_id.
self.docstore.add({_id: doc for _, _id, doc in full_info})
index_to_id = {index: _id for index, _id, _ in full_info}
self.index_to_docstore_id.update(index_to_id)
@classmethod
def __from(
cls,
texts: List[str],
embeddings: List[List[float]],
embedding: Embeddings,
metadatas: Optional[List[dict]] = None,
**kwargs: Any,
) -> FAISS:
faiss = dependable_faiss_import()
index = faiss.IndexFlatL2(len(embeddings[0]))
index.add(np.array(embeddings, dtype=np.float32))
documents = []
for i, text in enumerate(texts):
metadata = metadatas[i] if metadatas else {}
documents.append(Document(page_content=text, metadata=metadata))
index_to_id = {i: str(uuid.uuid4()) for i in range(len(documents))}
docstore = InMemoryDocstore(
{index_to_id[i]: doc for i, doc in enumerate(documents)}
)
return cls(embedding.embed_query, index, docstore, index_to_id)
@classmethod
def from_texts(
cls,
texts: List[str],
embedding: Embeddings,
metadatas: Optional[List[dict]] = None,
**kwargs: Any,
) -> FAISS:
"""Construct FAISS wrapper from raw documents.
This is a user friendly interface that:
1. Embeds documents.
2. Creates an in memory docstore
3. Initializes the FAISS database
This is intended to be a quick way to get started.
Example:
.. code-block:: python
from langchain import FAISS
from langchain.embeddings import OpenAIEmbeddings
embeddings = OpenAIEmbeddings()
faiss = FAISS.from_texts(texts, embeddings)
"""
embeddings = embedding.embed_documents(texts)
return cls.__from(texts, embeddings, embedding, metadatas, **kwargs)
@classmethod
def from_embeddings(
cls,
text_embeddings: List[Tuple[str, List[float]]],
embedding: Embeddings,
metadatas: Optional[List[dict]] = None,
**kwargs: Any,
) -> FAISS:
"""Construct FAISS wrapper from raw documents.
This is a user friendly interface that:
1. Embeds documents.
2. Creates an in memory docstore
3. Initializes the FAISS database
This is intended to be a quick way to get started.
Example:
.. code-block:: python
from langchain import FAISS
from langchain.embeddings import OpenAIEmbeddings
embeddings = OpenAIEmbeddings()
faiss = FAISS.from_texts(texts, embeddings)
"""
texts = [t[0] for t in text_embeddings]
embeddings = [t[1] for t in text_embeddings]
return cls.__from(texts, embeddings, embedding, metadatas, **kwargs)
def save_local(self, folder_path: str) -> None:
"""Save FAISS index, docstore, and index_to_docstore_id to disk.
Args:
folder_path: folder path to save index, docstore,
and index_to_docstore_id to.
"""
path = Path(folder_path)
path.mkdir(exist_ok=True, parents=True)
# save index separately since it is not picklable
faiss = dependable_faiss_import()
faiss.write_index(self.index, str(path / "index.faiss"))
# save docstore and index_to_docstore_id
with open(path / "index.pkl", "wb") as f:
pickle.dump((self.docstore, self.index_to_docstore_id), f)
@classmethod
def load_local(cls, folder_path: str, embeddings: Embeddings) -> FAISS:
"""Load FAISS index, docstore, and index_to_docstore_id to disk.
Args:
folder_path: folder path to load index, docstore,
and index_to_docstore_id from.
embeddings: Embeddings to use when generating queries
"""
path = Path(folder_path)
# load index separately since it is not picklable
faiss = dependable_faiss_import()
index = faiss.read_index(str(path / "index.faiss"))
# load docstore and index_to_docstore_id
with open(path / "index.pkl", "rb") as f:
docstore, index_to_docstore_id = pickle.load(f)
return cls(embeddings.embed_query, index, docstore, index_to_docstore_id)
|