dhQSSKJr SSKrSSKrSSKrSSKJr SSKJrJ r J r J r J r J r Jr SSKrSSKJr SSKJr SSKJr SSKJr SS KJrJr SS KJr SS KJr SS jr SS jr!"SS\5r"g)) annotationsN)Path)AnyCallableDictIterableListOptionalTuple)Document) Embeddings guard_import) VectorStore) AddableMixinDocstore)InMemoryDocstore)DistanceStrategyc vU[R"[RRUSSS9SS5-nU$)z!Normalize vectors to unit length.T)axiskeepdimsg-q=N)npcliplinalgnorm)xs ^/var/www/html/shao/venv/lib/python3.13/site-packages/langchain_community/vectorstores/scann.py normalizers1T:E4 HHA Hc[S5$)z5 Import `scann` if available, otherwise raise error. scannrr rdependable_scann_importr$s   r c\rSrSrSrSS\R S4SSjjrSSSjjrSSSjjr SSSjjr SSS jjr SSS jjr SS S jjr SS!S jjrSS"S jjr\S#S$Sjj5r\SS%Sjj5r\SS&Sjj5rS'S(Sjjr\S'SS.S)Sjjj5rS*SjrSS SjjrSrg)+ScaNN!a`ScaNN` vector store. To use, you should have the ``scann`` python package installed. Example: .. code-block:: python from langchain_community.embeddings import HuggingFaceEmbeddings from langchain_community.vectorstores import ScaNN model_name = "sentence-transformers/all-mpnet-base-v2" db = ScaNN.from_texts( ['foo', 'bar', 'barz', 'qux'], HuggingFaceEmbeddings(model_name=model_name)) db.similarity_search('foo?', k=1) NFc dXlX lX0lX@lXplXPlX`lXlg)z%Initialize with necessary components.N) embeddingindexdocstoreindex_to_docstore_iddistance_strategyoverride_relevance_score_fn _normalize_L2 _scann_config) selfr)r*r+r,relevance_score_fn normalize_L2r- scann_configs r__init__ScaNN.__init__3s1#  $8!!2+=())r c [UR[5(d[SURS35e[ S5e)NSIf trying to add texts, the underlying docstore should support adding items, which does notz(Updates are not available in ScaNN, yet.) isinstancer+r ValueErrorNotImplementedError)r1texts embeddings metadatasidskwargss r__add ScaNN.__addHsF$--66''+}}oY@ ""LMMr c tURR[U55nUR"X4X#S.UD6$)a4Run more texts through the embeddings and add to the vectorstore. Args: texts: Iterable of strings to add to the vectorstore. metadatas: Optional list of metadatas associated with the texts. ids: Optional list of unique IDs. Returns: List of ids from adding the texts into the vectorstore. r?r@)r)embed_documentslist _ScaNN__add)r1r=r?r@rAr>s r add_textsScaNN.add_textsWs5$^^33DK@ zz%TyTVTTr c [UR[5(d[SURS35e[ U6upVUR "XV4X#S.UD6$)aYRun 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. ids: Optional list of unique IDs. Returns: List of ids from adding the texts into the vectorstore. r8r9rE)r:r+rr;ziprH)r1text_embeddingsr?r@rAr=r>s radd_embeddingsScaNN.add_embeddingsls`$$--66''+}}oY@   1zz%TyTVTTr c [S5e)zDelete by vector ID or other criteria. Args: ids: List of ids to delete. **kwargs: Other keyword arguments that subclasses might use. Returns: Optional[bool]: True if deletion is successful, False otherwise, None if not implemented. z*Deletions are not available in ScaNN, yet.)r<)r1r@rAs rdelete ScaNN.deletes""NOOr c ^[R"U/[RS9nUR(a [ U5nUR R XccUOU5upx/n [US5HupU S:XaM URU n URRU 5m[T[5(d[SU ST35eUbUR5V Vs0sH upU [U[5(dU/OU_M" nn n[!U4SjUR555(aU R#TUSU 45 MMU R#TUSU 45 M UR%S5nUbvUR&[(R*[(R,4;a[.R0O[.R2nU VVs/sHunnU"UU5(dMUU4PM n nnU SU$s snn fs snnf) aReturn docs most similar to query. Args: embedding: Embedding vector to look up documents similar to. k: Number of Documents to return. Defaults to 4. filter (Optional[Dict[str, Any]]): Filter by metadata. Defaults to None. fetch_k: (Optional[int]) Number of Documents to fetch before filtering. Defaults to 20. **kwargs: kwargs to be passed to similarity search. Can include: score_threshold: Optional, a floating point value between 0 to 1 to filter the resulting set of retrieved docs Returns: List of documents most similar to the query text and L2 distance in float for each. Lower score represents more similarity. dtypeNrrzCould not find document for id z, got c3b># UH%upTRRU5U;v M' g7fN)metadataget).0keyvaluedocs r ?ScaNN.similarity_search_with_score_by_vector..s'W*#s||'',5s,/score_threshold)rarrayfloat32r/rr*search_batched enumerater,r+searchr:r r;itemsrGallappendrYr-rMAX_INNER_PRODUCTJACCARDoperatorgele)r1r)kfilterfetch_krAvectorindicesscoresdocsji_idr[r\r`cmpr] similaritys ` r&similarity_search_with_score_by_vector,ScaNN.similarity_search_with_score_by_vectors09+RZZ8   v&F**33 AW gaj)DABw++A.C--&&s+Cc8,, #B3%vcU!STT!'-lln&4  5$(?(?%UJ&4W WWWKKfQil 34X S&)A,/0!*$!**%67  &))$668H8P8PQR [[ (,'+OCz?3"j!'+  BQx-"s'&HH1Hc jURRU5nUR"UU4UUS.UD6nU$)aReturn docs most similar to query. Args: query: Text to look up documents similar to. k: Number of Documents to return. Defaults to 4. filter (Optional[Dict[str, str]]): Filter by metadata. Defaults to None. fetch_k: (Optional[int]) Number of Documents to fetch before filtering. Defaults to 20. Returns: List of documents most similar to the query text with L2 distance in float. Lower score represents more similarity. rorp)r) embed_queryrz)r1queryrnrorprAr)rts rsimilarity_search_with_score"ScaNN.similarity_search_with_scoresL*NN..u5 ::        r c hUR"UU4UUS.UD6nUVVs/sHupxUPM snn$s snnf)aReturn docs most similar to embedding vector. Args: embedding: Embedding to look up documents similar to. k: Number of Documents to return. Defaults to 4. filter (Optional[Dict[str, str]]): Filter by metadata. Defaults to None. fetch_k: (Optional[int]) Number of Documents to fetch before filtering. Defaults to 20. Returns: List of Documents most similar to the embedding. r})rz) r1r)rnrorprAdocs_and_scoresr]_s rsimilarity_search_by_vector!ScaNN.similarity_search_by_vectorsM(EE       #22//222s .c dUR"X4X4S.UD6nUVVs/sHupxUPM snn$s snnf)aReturn docs most similar to query. Args: query: Text to look up documents similar to. k: Number of Documents to return. Defaults to 4. filter: (Optional[Dict[str, str]]): Filter by metadata. Defaults to None. fetch_k: (Optional[int]) Number of Documents to fetch before filtering. Defaults to 20. Returns: List of Documents most similar to the query. r})r) r1rrnrorprArr]rs rsimilarity_searchScaNN.similarity_searchsC(;;  # 8> #22//222s ,c ([S5nURS[R5n URSS5n [R "U[R S9n U(a [U 5n U bURRX5n OU [R:Xa:URRU SS5R5R5n O9URRU SS5R5R5n /n Uc-UVs/sH!n[[R "55PM# nn[#U5H,unnU(aXOO0nU R%['UUS95 M. [)[#U55n[+U5[+U 5:wa#[-[+U5S [+U 5S 35e[/[)[1UR35U 555nU"UU UU4S U0UD6$s snf) Nr"r-r4rT dot_product squared_l2) page_contentrXz ids provided for z, documents. Each document should have an id.r3)rrYrEUCLIDEAN_DISTANCErrarbrscann_ops_pybindcreate_searcherribuilderscore_brute_forcebuildstruuiduuid4rdrhr dictlen ExceptionrrLvalues)clsr=r>r)r?r@r3rAr"r-r4rqr* documentsrrvtextrX index_to_idr+s r__from ScaNN.__from+sW%"JJ !1!D!D zz.$7 *BJJ7 v&F  #**::6PE $4$F$FF**2261mL&&(UW**2261lK&&(UW  ;.34e3tzz|$eC4 'GAt'0y|bH   X4(K L(9S>* { s9~ -{#$$6s9~6FG44  $D[-?-?-A9)M$NO      &     5s"'Hc TURU5nUR"UUU4UUS.UD6$)aConstruct ScaNN wrapper from raw documents. This is a user friendly interface that: 1. Embeds documents. 2. Creates an in memory docstore 3. Initializes the ScaNN database This is intended to be a quick way to get started. Example: .. code-block:: python from langchain_community.vectorstores import ScaNN from langchain_community.embeddings import OpenAIEmbeddings embeddings = OpenAIEmbeddings() scann = ScaNN.from_texts(texts, embeddings) rE)rF _ScaNN__from)rr=r)r?r@rAr>s r from_textsScaNN.from_textsgsE4..u5 zz          r c UVs/sHofSPM nnUVs/sHofSPM nnUR"UUU4UUS.UD6$s snfs snf)aConstruct ScaNN wrapper from raw documents. This is a user friendly interface that: 1. Embeds documents. 2. Creates an in memory docstore 3. Initializes the ScaNN database This is intended to be a quick way to get started. Example: .. code-block:: python from langchain_community.vectorstores import ScaNN from langchain_community.embeddings import OpenAIEmbeddings embeddings = OpenAIEmbeddings() text_embeddings = embeddings.embed_documents(texts) text_embedding_pairs = list(zip(texts, text_embeddings)) scann = ScaNN.from_embeddings(text_embedding_pairs, embeddings) rrrE)r) rrMr)r?r@rAtr=r>s rfrom_embeddingsScaNN.from_embeddingssi8 //!1/$34OqdO 4zz          04s ? Acl[U5nUSRUS9- nURSSS9 URR [ U55 [ USRUS9- S5n[R"URUR4U5 SSS5 g!,(df  g=f)zSave ScaNN index, docstore, and index_to_docstore_id to disk. Args: folder_path: folder path to save index, docstore, and index_to_docstore_id to. {index_name}.scann index_nameTexist_okparents{index_name}.pklwbN) rformatmkdirr* serializeropenpickledumpr+r,)r1 folder_pathrpath scann_pathfs r save_localScaNN.save_localsK 077:7NN $5 S_-$+22j2II4 PTU KK(A(ABA FQ P Ps ..B%% B3)allow_dangerous_deserializationc U(d [S5e[U5nUSRUS9- nURSSS9 [ S5nUR R [U55n [USRUS9- S5n [R"U 5un n S S S 5 U"X)W W 40UD6$!,(df  N=f) a=Load ScaNN index, docstore, and index_to_docstore_id from disk. Args: folder_path: folder path to load index, docstore, and index_to_docstore_id from. embedding: Embeddings to use when generating queries index_name: for saving with a specific index file name allow_dangerous_deserialization: whether to allow deserialization of the data which involves loading a pickle file. Pickle files can be modified by malicious actors to deliver a malicious payload that results in execution of arbitrary code on your machine. aBThe de-serialization relies loading a pickle file. Pickle files can be modified to deliver a malicious payload that results in execution of arbitrary code on your machine.You will need to set `allow_dangerous_deserialization` to `True` to enable deserialization. If you do this, make sure that you trust the source of the data. For example, if you are loading a file that you created, and know that no one else has modified the file, then this is safe to do. Do not set this to `True` if you are loading a file from an untrusted source (e.g., some random site on the internet.).rrTrr"rrbN) r;rrrrr load_searcherrrrload) rrr)rrrArrr"r*rr+r,s r load_localScaNN.load_locals./ "  K 077:7NN $5W%&&44S_E$+22j2II4 PTU  $Q9X/CNvNNQ Ps B88 CcURb UR$UR[R:Xa UR$UR[R :Xa UR $[S5e)a The 'correct' relevance function may differ depending on a few things, including: - the distance / similarity metric used by the VectorStore - the scale of your embeddings (OpenAI's are unit normed. Many others are not!) - embedding dimensionality - etc. zJUnknown distance strategy, must be cosine, max_inner_product, or euclidean)r.r-rri%_max_inner_product_relevance_score_fnr_euclidean_relevance_score_fnr;)r1s r_select_relevance_score_fn ScaNN._select_relevance_score_fnsr  + + 733 3  ! !%5%G%G G== =  # #'7'J'J J55 5  r c $URSS5nUR5nUc [S5eUR"U4UUUS.UD6nUV V s/sHupX"U 54PM n n n UbU V V s/sHupX:dM X4PM n n n U $s sn n fs sn n f)z?Return docs and their similarity scores on a scale from 0 to 1.r`NzLnormalize_score_fn must be provided to ScaNN constructor to normalize scores)rnrorp)poprr;r) r1rrnrorprAr`r2rr]scoredocs_and_rel_scoresrys r(_similarity_search_with_relevance_scores.ScaNN._similarity_search_with_relevance_scoress!**%6=!<<>  %9 ;;      @O ?NS$U+ ,    &(;#':OC0"!': # #" #s B, B :B )r/r0r-r+r)r*r,r.)r)r r*rr+rr,zDict[int, str]r2z"Optional[Callable[[float], float]]r3boolr-rr4z Optional[str])NN) r= Iterable[str]r>zIterable[List[float]]r?Optional[List[dict]]r@Optional[List[str]]rArreturn List[str]) r=rr?rr@rrArrr) rMz!Iterable[Tuple[str, List[float]]]r?rr@rrArrrrW)r@rrArrzOptional[bool])N) r) List[float]rnintroOptional[Dict[str, Any]]rprrArrList[Tuple[Document, float]]) rrrnrrorrprrArrr) r)rrnrrorrprrArrList[Document]) rrrnrrorrprrArrr)NNF)r=rr>zList[List[float]]r)r r?rr@rr3rrArrr&) r=rr)r r?rr@rrArrr&) rMzList[Tuple[str, List[float]]]r)r r?rr@rrArrr&)r*)rrrrrNone) rrr)r rrrrrArrr&)rzCallable[[float], float])__name__ __module__ __qualname____firstlineno____doc__rrr5rHrIrNrQrzrrr classmethodrrrrrrr__static_attributes__r#r rr&r&!sR.BF".>.Q.Q&**** * - * ? **,*$*2+/#' N N* N( N ! N  N  N$+/#' UU(U! U  U  U0+/#' U:U(U! U  U  U8 P"+/ >> >) >  >  > &>F+/  )      &D+/ 33 3) 3  3  3 3@+/ 33 3) 3  3  3 32 +/#'"9 9 &9  9 ( 9 ! 9 9 9  9 9 v +/#' ! ! ! ( ! ! !  !  ! ! F +/#' $ 6$ $ ( $ ! $  $  $ $ LG$ " 3O 16 3O3O3O 3O *. 3O3O 3O3Oj8+/ "#"# "#) "#  "#  "# &"#"#r r&)r np.ndarrayrr)rr)# __future__rrkrrpathlibrtypingrrrrr r r numpyrlangchain_core.documentsr langchain_core.embeddingsr langchain_core.utilsrlangchain_core.vectorstoresr!langchain_community.docstore.baserr&langchain_community.docstore.in_memoryr&langchain_community.vectorstores.utilsrrr$r&r#r rrsM" GGG-0-3DCC !T#KT#r