BTh}ddlmZddlZddlZddlZddlZddlmZmZm Z m Z m Z m Z m Z mZddlZddlmZddlmZddlmZddlmZddlmZdd lmZdd lmZ dd l m!Z!dd l m"Z#dd l$m%Z%ddl)m*Z*m+Z+er ejXe-Z.edeZ/GddeZ0edddGdde0Zy#e&$rddl'm#Z#m%Z%dd l(m!Z!YawxYw)) annotationsN) TYPE_CHECKINGAnyCallableIterableListOptionalTupleTypeVar) deprecated)Document) Embeddings) batch_iterate) VectorStore)Pinecone)PineconeAsyncio) ApplyResult)Index) _IndexAsyncio)_Indexr)DistanceStrategymaximal_marginal_relevanceVST)boundcLeZdZUdZdZded<dZded<ddddejfddd d%d Z e d&d Z e d'd Z e d(d Z d)dd  d*dZ d)dd  d+dZ d, d-dZ d, d.dZdddd d/dZdddd d/dZ d, d0dZ d, d1dZd2dZed3dZ d4 d5dZ d4 d5dZ d4 d6dZ d4 d7dZe d8dd d9dZe d:dd  d;d Ze d d?d"Z! d@ dAd#Z" d@ dAd$Z#y)BPineconeVectorStoreuoPinecone vector store integration. Setup: Install ``langchain-pinecone`` and set the environment variable ``PINECONE_API_KEY``. .. code-block:: bash pip install -qU langchain-pinecone export PINECONE_API_KEY = "your-pinecone-api-key" Key init args — indexing params: embedding: Embeddings Embedding function to use. Key init args — client params: index: Optional[Index] Index to use. # TODO: Replace with relevant init params. Instantiate: .. code-block:: python import time import os from pinecone import Pinecone, ServerlessSpec from langchain_pinecone import PineconeVectorStore from langchain_openai import OpenAIEmbeddings pc = Pinecone(api_key=os.environ.get("PINECONE_API_KEY")) index_name = "langchain-test-index" # change if desired existing_indexes = [index_info["name"] for index_info in pc.list_indexes()] if index_name not in existing_indexes: pc.create_index( name=index_name, dimension=1536, metric="cosine", spec=ServerlessSpec(cloud="aws", region="us-east-1"), deletion_protection="enabled", # Defaults to "disabled" ) while not pc.describe_index(index_name).status["ready"]: time.sleep(1) index = pc.Index(index_name) vector_store = PineconeVectorStore(index=index, embedding=OpenAIEmbeddings()) Add Documents: .. code-block:: python from langchain_core.documents import Document document_1 = Document(page_content="foo", metadata={"baz": "bar"}) document_2 = Document(page_content="thud", metadata={"bar": "baz"}) document_3 = Document(page_content="i will be deleted :(") documents = [document_1, document_2, document_3] ids = ["1", "2", "3"] vector_store.add_documents(documents=documents, ids=ids) Delete Documents: .. code-block:: python vector_store.delete(ids=["3"]) Search: .. code-block:: python results = vector_store.similarity_search(query="thud",k=1) for doc in results: print(f"* {doc.page_content} [{doc.metadata}]") .. code-block:: python * thud [{'bar': 'baz'}] Search with filter: .. code-block:: python results = vector_store.similarity_search(query="thud",k=1,filter={"bar": "baz"}) for doc in results: print(f"* {doc.page_content} [{doc.metadata}]") .. code-block:: python * thud [{'bar': 'baz'}] Search with score: .. code-block:: python results = vector_store.similarity_search_with_score(query="qux",k=1) for doc, score in results: print(f"* [SIM={score:3f}] {doc.page_content} [{doc.metadata}]") .. code-block:: python * [SIM=0.832268] foo [{'baz': 'bar'}] Async: .. code-block:: python # add documents # await vector_store.aadd_documents(documents=documents, ids=ids) # delete documents # await vector_store.adelete(ids=["3"]) # search # results = vector_store.asimilarity_search(query="thud",k=1) # search with score results = await vector_store.asimilarity_search_with_score(query="qux",k=1) for doc,score in results: print(f"* [SIM={score:3f}] {doc.page_content} [{doc.metadata}]") .. code-block:: python * [SIM=0.832268] foo [{'baz': 'bar'}] Use as Retriever: .. code-block:: python retriever = vector_store.as_retriever( search_type="mmr", search_kwargs={"k": 1, "fetch_k": 2, "lambda_mult": 0.5}, ) retriever.invoke("thud") .. code-block:: python [Document(metadata={'bar': 'baz'}, page_content='thud')] NzOptional[_Index]_indexzOptional[_IndexAsyncio] _async_indextext)pinecone_api_key index_namecT| td||_| td||_||_||_|rVt |t r||_n||_|jj|_ |jj|_ y|xs#tjj!dxsd}|s td||_ |xs#tjj!dxsd} | s td||_ t#|d } | j%| |_y) NzEmbedding must be providedzText key must be providedPINECONE_API_KEYziPinecone API key must be provided in either `pinecone_api_key` or `PINECONE_API_KEY` environment variablePINECONE_INDEX_NAMEziPinecone index name must be provided in either `index_name` or `PINECONE_INDEX_NAME` environment variable langchainapi_key source_tag)name) ValueError _embedding _text_key _namespacedistance_strategy isinstancerrrconfighost _index_hostr(_pinecone_api_keyosenvirongetPineconeClientr) selfindex embeddingtext_key namespacer/r r!r4 _index_nameclients q/var/www/catia.catastroantioquia-mas.com/valormas/lib/python3.12/site-packages/langchain_pinecone/vectorstores.py__init__zPineconeVectorStore.__init__s&  9: :#  89 9!#!2 %/$)!# $||00D %*\\%9%9D "!LBJJNN3E$FL" % A&7D "$S 7L(MSQSK D&7D "#,=+VF ,,K,8DKc|jJt|ds tdt|jd}|j |j S|jS)zGet synchronous index instance.r4zNo Pinecone API key availabler&r'r2)rhasattrr+r8r4rr3r9r?s r@r:zPineconeVectorStore.indexs] ;; 4!45 !@AA#..;F<4 when constructing the pinecone.Index, embedding_chunk_size>1000 and batch_size~64 for best performance. Args: texts: Iterable of strings to add to the vectorstore. metadatas: Optional list of metadatas associated with the texts. ids: Optional list of ids to associate with the texts. namespace: Optional pinecone namespace to add the texts to. batch_size: Batch size to use when adding the texts to the vectorstore. embedding_chunk_size: Chunk size to use when embedding the texts. id_prefix: Optional string to use as an ID prefix when upserting vectors. Returns: List of ids from adding the texts into the vectorstore. #rvectorsr=) r.liststruuiduuid4zipr-rangelenr,embed_documentsr:upsert)r9texts metadatasidsr= batch_sizeembedding_chunk_sizerNkwargs_idmetadatari chunk_texts chunk_idschunk_metadatasrM vector_tupless r@ add_textszPineconeVectorStore.add_texts sl>  IU 77Ac$**,'7 TWNP Cr(A C"$rIC4e!4"!4 !)U3 ,NHd'+HT^^ $ , q#e*&:; AA(<$<=KA$8 89I'A0D,DEO88EJ Y O!LMM DJJ   %#    /8"5s$D%D*1 D/c ZK| |j}t|}|xs+|D cgc]} ttj!c} }|r|D cgc]} |dz| vr|dz| zn| }} |xs|D cgc]} ic} }t ||D]\} } | | |j <tdt||D]} || | |z}|| | |z}|| | |z}|jj|d{}t |||}|j4d{}g}t||D](}|jd||d|}|j|*tj |d{dddd{|Scc} wcc} wcc} w77|7+7#1d{7swYxYww)aAsynchronously run more texts through the embeddings and add to the vectorstore. Upsert optimization is done by chunking the embeddings and upserting them. This is done to avoid memory issues and optimize using HTTP based embeddings. For OpenAI embeddings, use pool_threads>4 when constructing the pinecone.Index, embedding_chunk_size>1000 and batch_size~64 for best performance. Args: texts: Iterable of strings to add to the vectorstore. metadatas: Optional list of metadatas associated with the texts. ids: Optional list of ids to associate with the texts. namespace: Optional pinecone namespace to add the texts to. batch_size: Batch size to use when adding the texts to the vectorstore. embedding_chunk_size: Chunk size to use when embedding the texts. id_prefix: Optional string to use as an ID prefix when upserting vectors. Returns: List of ids from adding the texts into the vectorstore. NrPrrQrS)r.rTrUrVrWrXr-rYrZr,aembed_documentsrJrr\appendasynciogather)r9r]r^r_r=r`rarNrbrcrdrerrfrgrhrirMrjidxtasksbatch_vector_tuplestasks r@ aadd_textszPineconeVectorStore.aadd_textsIs>  IU 77Ac$**,'7 TWNP Cr(A C"$rIC4e!4"!4 !)U3 ,NHd'+HT^^ $ ,q#e*&:; -AA(<$<=KA$8 89I'A0D,DEO#?? LLJ :GM'' - -3+8]+S''%3:: 3"+!D LL& 'nne,,, - - - -* ?8"5M -- - - - -s"F+$E? F+F) F+3 F >NO  rBc K| |j}g}|j4d{}|j||d||d{}dddd{dD]}|d} |jd} |j| vr@| j |j} |d} |j t| | | | fgtjd |jd |S777#1d{7swYxYww) zVReturn pinecone documents most similar to embedding, along with scores asynchronously.NTrrrerdrrrr) r.rJr~r7r-rrnr rr) r9r;ryrzr=rrqrrrerdrrs r@rz;PineconeVectorStore.asimilarity_search_by_vector_with_scores  I##  sII !%# &G  9% C:HB~~)||DNN3G  $JER.t~~.>>NO  -     sW!DC)DC/C+C/ DC-BD+C/-D/D5C8 6D=Dc f|j|f|||d|}|Dcgc]\}}| c}}Scc}}w)aReturn pinecone documents most similar to query. Args: query: Text to look up documents similar to. k: Number of Documents to return. Defaults to 4. filter: Dictionary of argument(s) to filter on metadata namespace: Namespace to search in. Default will search in '' namespace. Returns: List of Documents most similar to the query and score for each rx)r r9r~ryrzr=rbdocs_and_scoresdocrcs r@similarity_searchz%PineconeVectorStore.similarity_searchsG&<$;;  v >D #22Q222s -cK|j|f|||d|d{}|Dcgc]\}}| c}}S7cc}}ww)Nrx)rrs r@asimilarity_searchz&PineconeVectorStore.asimilarity_searchsX!C B B ! v! >D!  #22Q22 3s?7 ? 9??c|jtjk(r |jS|jtjk(r |j S|jtj k(r |jStd)a8 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. zXUnknown distance strategy, must be cosine, max_inner_product (dot product), or euclidean) r/rCOSINE_cosine_relevance_score_fnMAX_INNER_PRODUCT%_max_inner_product_relevance_score_fnEUCLIDEAN_DISTANCE_euclidean_relevance_score_fnr+rLs r@_select_relevance_score_fnz.PineconeVectorStore._select_relevance_score_fn,sy  ! !%5%<%< <22 2  # #'7'I'I I== =  # #'7'J'J J55 5. rBc|dzdz S)z8Pinecone returns cosine similarity scores between [-1,1]rS)rs r@rz.PineconeVectorStore._cosine_relevance_score_fnBs QrBc | |j}|jj||dd||}t|tr t dt tj|gtj|dD cgc]} | d c} ||} | D cgc] } |d| d} } | D cgc](} t| j|j| *c} Scc} wcc} wcc} w) aReturn 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. lambda_mult: Number between 0 and 1 that determines the degree of diversity among the results with 0 corresponding to maximum diversity and 1 to minimum diversity. Defaults to 0.5. filter: Dictionary of argument(s) to filter on metadata namespace: Namespace to search in. Default will search in '' namespace. Returns: List of Documents selected by maximal marginal relevance. Trrinclude_valuesrr=rzrdtypervaluesry lambda_multrerre) r.r:r~r0rr+rnparrayfloat32r rr-)r9r;ryfetch_krrzr=rbritem mmr_selectedrfselectedres r@'max_marginal_relevance_search_by_vectorz;PineconeVectorStore.max_marginal_relevance_search_by_vectorGs:  I**""! #  g{ +RS S1 HHi[ 3(/ (: ;T(^ ;#  @LL!GI&q)*5LL%  (,,"@8 T  <M s< C C#.-C(c (K| |j}|j4d{}|j||dd||d{} dddd{tt j |gtj  dD cgc]} | d c} ||} | D cgc] } | d| d} } | Dcgc](}t|j|j| *c}S777#1d{7swYxYwcc} wcc} wcc}ww) aReturn docs selected using the maximal marginal relevance asynchronously. 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. lambda_mult: Number between 0 and 1 that determines the degree of diversity among the results with 0 corresponding to maximum diversity and 1 to minimum diversity. Defaults to 0.5. filter: Dictionary of argument(s) to filter on metadata namespace: Namespace to search in. Default will search in '' namespace. Returns: List of Documents selected by maximal marginal relevance. NTrrrrrrer) r.rJr~rrrrr rr-)r9r;ryrrrzr=rbrqrrrrfrres r@(amax_marginal_relevance_search_by_vectorz R  #      <M szDC(DC.C*C. DC,5D D D D2D8-D %D*C.,D.D4C7 5D<Dc d|jj|}|j||||||S)aReturn 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. lambda_mult: Number between 0 and 1 that determines the degree of diversity among the results with 0 corresponding to maximum diversity and 1 to minimum diversity. Defaults to 0.5. filter: Dictionary of argument(s) to filter on metadata namespace: Namespace to search in. Default will search in '' namespace. Returns: List of Documents selected by maximal marginal relevance. )r,r|r r9r~ryrrrzr=rbr;s r@max_marginal_relevance_searchz1PineconeVectorStore.max_marginal_relevance_searchs8:OO//6 ;; q';   rBcK|jj|d{}|j||||||d{S7#7w)N)ryrrrzr=)r,rrrs r@amax_marginal_relevance_searchz2PineconeVectorStore.amax_marginal_relevance_searchsZ//66u== BB # C   > s!A AA AA A )r c|xs#tjjdxsd}t||d}|j }|j dDcgc]}|j }}|r||vr|j|} | St|dk(r tdtd|d d j|cc}w) aReturn a Pinecone Index instance. Args: index_name: Name of the index to use. pool_threads: Number of threads to use for index upsert. pinecone_api_key: The api_key of Pinecone. Returns: Pinecone Index instance.r#r$r&)r( pool_threadsr)indexesrzNo active indexes found in your Pinecone project, are you sure you're using the right Pinecone API key and Environment? Please double check your Pinecone dashboard.zIndex 'zQ' not found in your Pinecone project. Did you mean one of the following indexes: z, ) r5r6r7r8 list_indexes index_listr*rrZr+join) clsr!rr r4r?rrf index_namesr:s r@get_pinecone_indexz&PineconeVectorStore.get_pinecone_indexs -X ?Q0RXVX%L[ %%''.'9'9)'DE!qvvE E * 3LL,E   "?  *&>>Bii >T=UW FsB>c  ~|j|| }|||||fi| }|j|f||||| | d| xsi|S)aConstruct Pinecone wrapper from raw documents. This is a user-friendly interface that: 1. Embeds documents. 2. Adds the documents to a provided Pinecone index This is intended to be a quick way to get started. The `pool_threads` affects the speed of the upsert operations. Setup: set the `PINECONE_API_KEY` environment variable to your Pinecone API key. Example: .. code-block:: python from langchain_pinecone import PineconeVectorStore, PineconeEmbeddings embeddings = PineconeEmbeddings(model="multilingual-e5-large") index_name = "my-index" vectorstore = PineconeVectorStore.from_texts( texts, index_name=index_name, embedding=embedding, namespace=namespace, ) r^r_r=r`rarN)rrk)rr]r;r^r_r`r<r=r! upsert_kwargsrembeddings_chunk_sizerNrbpinecone_indexpinecones r@ from_textszPineconeVectorStore.from_textsslZ// LI~y(IPP  !!6 " rBc  xK|d||||d| } | j|f||||| | d| xsid{| S7w)N)r!r;r<r=rrS)ru)rr]r;r^r_r`r<r=r!rrrNrbrs r@ afrom_textszPineconeVectorStore.afrom_textsJs~" !    "h!!   !!6  "   s /:8:c<|j||}|||||S)z*Load pinecone vectorstore from index name.)r)rr!r;r<r=rrs r@from_existing_indexz'PineconeVectorStore.from_existing_indexps'// LI>9h BBrBc J| |j}|r |jjd d|d|y|Ed}tdt ||D])}||||z}|jjd ||d|+y| |jjd ||d|yt d) a=Delete by vector IDs or filter. Args: ids: List of ids to delete. delete_all: Whether delete all vectors in the index. filter: Dictionary of conditions to filter vectors to delete. namespace: Namespace to search in. Default will search in '' namespace. NT delete_allr=rr_r=rzr=3Either ids, delete_all, or filter must be provided.rS)r.r:deleterYrZr+) r9r_rr=rzrb chunk_sizerfchunks r@rzPineconeVectorStore.delete}s  I  DJJ   M Mf M_J1c#h 3 LAJ/! !!KeyKFK L   DJJ   KVy KF KRS SrBc K| |j}|rC|j4d{}|jdd|d|d{dddd{y|d}|j4d{}g}tdt ||D].} || | |z} |j |jd| |d|0t j|d{dddd{y|C|j4d{}|jd||d|d{dddd{ytd777#1d{7swYyxYw777u#1d{7swYyxYw7t7Y7K#1d{7swYyxYww) NTrrrrrrrS) r.rJrrYrZrnrorpr+) r9r_rr=rzrbrqrrrrfrs r@adeletezPineconeVectorStore.adeletes  I '' Q Q3 cjjPDIPPPP Q Q _J'' - -3q#c(J7WAA N3ELL!U!Uf!UVWnne,,,  - -  '' O O3 cjjN)NvNNN O O RS S QP Q Q Q Q  - -  - - - -  ON O O O O s!FD3FD:D6D: FD8F*E+F.AE EE FEF5E*6F9E0E,E0 F"E.#F6D:8F:E E E FEFE'E E'#F,E0.F0F6E9 7F>F)r:z Optional[Any]r;Optional[Embeddings]r< Optional[str]r=rr/zOptional[DistanceStrategy]r rr!r)returnr)rr)rr)NNN r)r] Iterable[str]r^Optional[List[dict]]r_Optional[List[str]]r=rr`intrarrNrrbrr List[str])r]rr^rr_rr=rr`rrarrNrrbrrz list[str])rvNN) r~rUryrrzOptional[dict]r=rrList[Tuple[Document, float]]) r~rUryrrzrr=rrzlist[tuple[Document, float]]) r; List[float]ryrrzrr=rrr) r~rUryrrzrr=rrbrrList[Document]) r~rUryrrzrr=rrbrrlist[Document])rzCallable[[float], float])rfloatrr)rvg?NN)r;rryrrrrrrzrr=rrbrrr)r~rUryrrrrrrzrr=rrbrrr)r~rUryrrrrrrzrr=rrbrrr)rv)r!rrrr rrr) NNrrNNNrvr)r]rr;rr^rr_rr`rr<rUr=rr!rrrrrrrrNrrbrrr)NNrrNNNr)rztype[PineconeVectorStore]r]rr;rr^rr_rr`rr<rUr=rr!rrrrrrNrrbrrr)rNrv) r!rUr;rr<rUr=rrrrr)NNNN) r_rrzOptional[bool]r=rrzrrbrrNone)$__name__ __module__ __qualname____doc__r__annotations__rrrrApropertyr:rJrMrkrurrr{rrrr staticmethodrrrrr classmethodrrrrrrrSrBr@rr.sFP $F #,0L)0 $*."(#'8H8O8O69+/$(69  69(69 69!69669(69"69p  !!+/#'#'$(:$(::(:! : ! :  :":!:: :~+/#'#'$(B$(BB(B! B ! B  B"B!BB BN!%#'       !  &  2!%#'       !  &  :!%#' $$  $  $ ! $ &$T!%#' ##  #  # ! # &#P!%#' 33 3 3 ! 3  3 36!%#' 3 3  3 3 ! 3  3  3, !%#'3 3  3  3  3  3 !3 3  3 p !%#'4 4  4  4  4  4 !4 4  4 r !%#'             !       J !%#'          !    (# +/ #!## ( #  ##J +/#'#'$((,%)9$(999( 9 ! 9  99!9"9&99 #9!99 !99v +/#'#'$((,%)#$(# &###( # ! #  ##!#"#&# ##!## ##J #' C C C C ! C  C  C C$(%)#'!%  #!      F$(%)#'!%  #!      rBrz0.0.3z1.0.0)sinceremoval alternativeceZdZdZy)rz,Deprecated. Use PineconeVectorStore instead.N)rrrrrSrBr@rrs6rBr)1 __future__rrologgingr5rVtypingrrrrrr r r numpyrlangchain_core._api.deprecationr langchain_core.documentsr langchain_core.embeddingsrlangchain_core.utils.iterrlangchain_core.vectorstoresrrrr8rrHpinecone.db_data.indexrrrpinecone.db_data.index_asyncior ImportError pinecone.datapinecone.data.indexlangchain_pinecone._utilitiesrr getLoggerrrrrrSrBr@rs"    6-033/=026< W   8 $ e;'L+L^ '78MN " O }03/0sB88C C