a bgm#@sddlmZddlZddlZddlmZmZmZmZm Z ddl m Z ddl m Z ddlmZddlmZeeZGdd d eZdS) ) annotationsN)AnyIterableListOptionalTuple)Document) Embeddings) VectorStore)DistanceStrategyc @seZdZdZejfddddddZedd d d Zd d dddZ dddddZ d7dddddddZ d8d ddd ddd!d"d#Z d9d$d ddd%d&d'Z d(gfdd d)dd*d+d,d-Zd(gd.dd d)dd*d/d0d1Zd(gfdd d)dd$d+d2d3Zed:ddddddd4d5d6ZdS);KDBAIaT`KDB.AI` vector store. See https://kdb.ai. To use, you should have the `kdbai_client` python package installed. Args: table: kdbai_client.Table object to use as storage, embedding: Any embedding function implementing `langchain.embeddings.base.Embeddings` interface, distance_strategy: One option from DistanceStrategy.EUCLIDEAN_DISTANCE, DistanceStrategy.DOT_PRODUCT or DistanceStrategy.COSINE. See the example [notebook](https://github.com/KxSystems/langchain/blob/KDB.AI/docs/docs/integrations/vectorstores/kdbai.ipynb). rr zOptional[DistanceStrategy])table embeddingdistance_strategycCs>z ddl}Wnty&tdYn0||_||_||_dS)Nrz`Could not import kdbai_client python package. Please install it with `pip install kdbai_client`.) kdbai_client ImportError_table _embeddingr)selfr rrrrt/var/www/html/cobodadashboardai.evdpl.com/venv/lib/python3.9/site-packages/langchain_community/vectorstores/kdbai.py__init__!s   zKDBAI.__init__zOptional[Embeddings])returncCst|jtr|jSdSN) isinstancerr rrrr embeddings4s zKDBAI.embeddingsz Iterable[str]zList[List[float]])textsrcs.tjtrjt|Sfdd|DS)Ncsg|]}|qSr)r.0trrr =z*KDBAI._embed_documents..)rrr embed_documentslist)rrrrr_embed_documents:s zKDBAI._embed_documentsstrz List[float])textrcCs"t|jtr|j|S||Sr)rrr Z embed_query)rr'rrr _embed_query?s  zKDBAI._embed_queryNz List[str]zOptional[List[str]]z Optional[Any]None)ridsmetadatarcsz ddlWnty&tdYn0z ddl}WntyNtdYn0|j|}|}||d<dd|D|d<fdd|D|d <|dur|j||gd d }|jj|d d dS)NrzRCould not import numpy python package. Please install it with `pip install numpy`.TCould not import pandas python package. Please install it with `pip install pandas`.idcSsg|]}|dqS)zutf-8)encoderrrrr!]r"z!KDBAI._insert..r'csg|]}j|ddqS)Zfloat32)Zdtype)array)renprrr!^r"r)ZaxisF)warn) numpyrpandasrr# DataFrameconcatrinsert)rrr*r+pdZembedsZdfrr1r_insertDs(       z KDBAI._insert zOptional[List[dict]]int)r metadatasr* batch_sizekwargsrcKsz ddl}Wnty&tdYn0t|}d}|durXt||jrN|}n ||}g}t|d|d} t| D]} | |} | d|} || | } |r|| | }nddtt| D}|dur|j| | jdd}nd}| | ||||}qx|S) aRun more texts through the embeddings and add to the vectorstore. Args: texts (Iterable[str]): Texts to add to the vectorstore. metadatas (Optional[List[dict]]): List of metadata corresponding to each chunk of text. ids (Optional[List[str]]): List of IDs corresponding to each chunk of text. batch_size (Optional[int]): Size of batch of chunks of text to insert at once. Returns: List[str]: List of IDs of the added texts. rNr,r3cSsg|]}ttqSr)r&uuiduuid4)r_rrrr!r"z#KDBAI.add_texts..T)Zdrop) r6rr$rr7lenrangeZilocZ reset_indexr;)rrr>r*r?r@r:ZmetadfZout_idsZnbatchesiistartZiendbatchZ batch_idsZ batch_metarrr add_textscs6         zKDBAI.add_textszList[Document]) documentsr?r@rcKsZz ddl}Wnty&tdYn0dd|D}|dd|D}|j|||dS)aHRun more documents through the embeddings and add to the vectorstore. Args: documents (List[Document]: Documents to add to the vectorstore. batch_size (Optional[int]): Size of batch of documents to insert at once. Returns: List[str]: List of IDs of the added texts. rNr,cSsg|] }|jqSr) page_contentrxrrrr!r"z'KDBAI.add_documents..cSsg|] }|jqSr)r+rLrrrr!r")r+r?)r6rr7rI)rrJr?r@r:rr+rrr add_documentss    zKDBAI.add_documentsr3zOptional[List]zList[Tuple[Document, float]])querykfilterr@rcKs|j||f||d|S)anRun similarity search with distance from a query string. Args: query (str): Query string. k (Optional[int]): number of neighbors to retrieve. filter (Optional[List]): KDB.AI metadata filter clause: https://code.kx.com/kdbai/use/filter.html Returns: List[Document]: List of similar documents. rPrQ)&similarity_search_by_vector_with_scorer()rrOrPrQr@rrrsimilarity_search_with_scoresz"KDBAI.similarity_search_with_scorerR)rrPrQr@rc Ksd|vr|d}|jjf|g||d|}g}t|trF|d}n|S|jddD]<}|d}|d} |t|dd |Dd | fqV|S) aReturn documents most similar to embedding, along with scores. Args: embedding (List[float]): query vector. k (Optional[int]): number of neighbors to retrieve. filter (Optional[List]): KDB.AI metadata filter clause: https://code.kx.com/kdbai/use/filter.html Returns: List[Document]: List of similar documents. n)ZvectorsrUrQrrecords)Zorientr'Z __nn_distancecSsi|]\}}|dkr||qS)r'r)rrPvrrr r"z@KDBAI.similarity_search_by_vector_with_score..)rKr+) poprsearchrr$to_dictappendritems) rrrPrQr@matchesdocsrowr'ZscorerrrrSs&      z,KDBAI.similarity_search_by_vector_with_scorecKs&|j|f||d|}dd|DS)a`Run similarity search from a query string. Args: query (str): Query string. k (Optional[int]): number of neighbors to retrieve. filter (Optional[List]): KDB.AI metadata filter clause: https://code.kx.com/kdbai/use/filter.html Returns: List[Document]: List of similar documents. rRcSsg|] \}}|qSrr)rdocrCrrrr!r"z+KDBAI.similarity_search..)rT)rrOrPrQr@Zdocs_and_scoresrrrsimilarity_searchszKDBAI.similarity_search)clsrrr>r@rcKs tddS)zNot implemented.N) Exception)rcrrr>r@rrr from_textss zKDBAI.from_texts)N)NNr<)r<)N)__name__ __module__ __qualname____doc__r ZEUCLIDEAN_DISTANCErpropertyrr%r(r;rIrNrTrSrb classmethodrerrrrr s6 "8+r ) __future__rloggingrAtypingrrrrrZlangchain_core.documentsrZlangchain_core.embeddingsr Zlangchain_core.vectorstoresr Z&langchain_community.vectorstores.utilsr getLoggerrfloggerr rrrrs