AhJ~SrSSKrSSKrSSKJr SSKJr SSKJrJ r SSK J r SSK J r Jr SSKJr SS KJr SS KJrJr SS KJr SS KJrJr SS KJr S\4Sjr\R<"\5r \RB"SS9S\"4Sj5r#S\4Sjr$"SS\%\5r&"SS\5r'"SS\'\5r("SS\'\5r)g)z@A chain for comparing the output of two models using embeddings.N)Enum)util)AnyOptional) Callbacks)AsyncCallbackManagerForChainRunCallbackManagerForChainRun) Embeddings)pre_init) ConfigDictField)Chain)PairwiseStringEvaluatorStringEvaluatorRUN_KEYreturncPSSKnU$![anSn[U5UeSnAff=f)Nrz@Could not import numpy, please install with `pip install numpy`.)numpy ImportError)npemsgs d/var/www/html/shao/venv/lib/python3.13/site-packages/langchain/evaluation/embedding_distance/base.py _import_numpyrs5& I &P#A%&s % %)maxsizecx[[R"S55(ag[R S5 g)NrTaNumPy not found in the current Python environment. langchain will use a pure Python implementation for embedding distance operations, which may significantly impact performance, especially for large datasets. For optimal speed and efficiency, consider installing NumPy: pip install numpyF)boolr find_specloggerwarningr _check_numpyr%#s/ DNN7 #$$ NN  r$cSSKJn U"5$![a0 SSKJn U"5$![anSn[U5UeSnAff=ff=f)zUCreate an Embeddings object. Returns: Embeddings: The created Embeddings object. rOpenAIEmbeddingstCould not import OpenAIEmbeddings. Please install the OpenAIEmbeddings package using `pip install langchain-openai`.N)langchain_openair(r%langchain_community.embeddings.openai)r(rrs r_embedding_factoryr,1sg *5    * *     *Q c" )  * *s# A ( AAAA c,\rSrSrSrSrSrSrSrSr Sr g ) EmbeddingDistanceIzEmbedding Distance Metric. Attributes: COSINE: Cosine distance metric. EUCLIDEAN: Euclidean distance metric. MANHATTAN: Manhattan distance metric. CHEBYSHEV: Chebyshev distance metric. HAMMING: Hamming distance metric. cosine euclidean manhattan chebyshevhammingr#N) __name__ __module__ __qualname____firstlineno____doc__COSINE EUCLIDEAN MANHATTAN CHEBYSHEVHAMMING__static_attributes__r#r$rr.r.Is"FIIIGr$r.c\rSrSr%Sr\"\S9r\\ S'\"\ RS9r \ \ S'\ S\\\4S\\\44S j5r\"S S 9r\S\\4S j5rS \S\4SjrS\ S\4Sjr\S\S\S\4Sj5r\S\S\S\4Sj5r\S\S\S\4Sj5r\S\S\S\4Sj5r\S\S\S\4Sj5rS\S\4Sjr Sr!g)_EmbeddingDistanceChainMixin[aShared functionality for embedding distance evaluators. Attributes: embeddings (Embeddings): The embedding objects to vectorize the outputs. distance_metric (EmbeddingDistance): The distance metric to use for comparing the embeddings. )default_factory embeddings)defaultdistance_metricvaluesrcvURS5n/nSSKJn URU5 SSKJn URU5 U(d Sn[ U5e[ U[U55(aSSKnU$U$![a N[f=f![a NSf=f![anSn[ U5UeSnAff=f)zValidate that the TikTok library is installed. Args: values (Dict[str, Any]): The values to validate. Returns: Dict[str, Any]: The validated values. rDrr'r)NzThe tiktoken library is required to use the default OpenAI embeddings with embedding distance evaluators. Please either manually select a different Embeddings object or install tiktoken using `pip install tiktoken`.) getr*r(appendrr+ isinstancetupletiktoken)clsrGrDtypes_r(rrMrs r_validate_tiktoken_installed9_EmbeddingDistanceChainMixin._validate_tiktoken_installedgsZZ -   9 MM* +   MM* +Q c" " j%- 0 0 . v =       .I "#&A- .s:A;B 3B; BB BB B8%B33B8T)arbitrary_types_allowedcS/$)zOReturn the output keys of the chain. Returns: List[str]: The output keys. scorer#selfs r output_keys(_EmbeddingDistanceChainMixin.output_keyss yr$resultcHSUS0n[U;aU[U['U$)NrTr)rVrYparseds r_prepare_output,_EmbeddingDistanceChainMixin._prepare_outputs*6'?+ f $WoF7O r$metricc <[RUR[RUR[R UR [RUR[RUR0nX;aX!$SU3n[U5e)zGet the metric function for the given metric name. Args: metric (EmbeddingDistance): The metric name. Returns: Any: The metric function. zInvalid metric: ) r.r:_cosine_distancer;_euclidean_distancer<_manhattan_distancer=_chebyshev_distancer>_hamming_distance ValueError)rVr^metricsrs r _get_metric(_EmbeddingDistanceChainMixin._get_metrics  $ $d&;&;  ' ')A)A  ' ')A)A  ' ')A)A  % %t'='=    ? " )or$abcfSSKJn SU"X5- $![anSn[U5UeSnAff=f)zCompute the cosine distance between two vectors. Args: a (np.ndarray): The first vector. b (np.ndarray): The second vector. Returns: np.ndarray: The cosine distance. r)cosine_similarityzThe cosine_similarity function is required to compute cosine distance. Please install the langchain-community package using `pip install langchain-community`.Ng?)langchain_community.utils.mathrlr)rirjrlrrs rr`-_EmbeddingDistanceChainMixin._cosine_distancesH * H&q,,, *6  c" )  *s 0+0c[5(a!SSKnURRX- 5$[ S[ X555S-$)zCompute the Euclidean distance between two vectors. Args: a (np.ndarray): The first vector. b (np.ndarray): The second vector. Returns: np.floating: The Euclidean distance. rNc36# UHupX- X- -v M g7fNr#.0xys r C_EmbeddingDistanceChainMixin._euclidean_distance..s;AEae$sg?)r%rlinalgnormsumziprirjrs rra0_EmbeddingDistanceChainMixin._euclidean_distances= >> 99>>!%( (;Q;;sBBr$c[5(a,[5nURURX- 55$[S[ X555$)zCompute the Manhattan distance between two vectors. Args: a (np.ndarray): The first vector. b (np.ndarray): The second vector. Returns: np.floating: The Manhattan distance. c3># UHup[X- 5v M g7frqabsrrs rrvC_EmbeddingDistanceChainMixin._manhattan_distance..4)$!3qu::))r%rrzrr{r|s rrb0_EmbeddingDistanceChainMixin._manhattan_distance@ >>B66"&&-( (4#a)444r$c[5(a,[5nURURX- 55$[S[ X555$)zCompute the Chebyshev distance between two vectors. Args: a (np.ndarray): The first vector. b (np.ndarray): The second vector. Returns: np.floating: The Chebyshev distance. c3># UHup[X- 5v M g7frqrrrs rrvC_EmbeddingDistanceChainMixin._chebyshev_distance..rr)r%rmaxrr{r|s rrc0_EmbeddingDistanceChainMixin._chebyshev_distancerr$c[5(a[5nURX:g5$[S[ X555[ U5- $)zCompute the Hamming distance between two vectors. Args: a (np.ndarray): The first vector. b (np.ndarray): The second vector. Returns: np.floating: The Hamming distance. c36# UHupX:wdM Sv M g7f)rNr#rrs rrvA_EmbeddingDistanceChainMixin._hamming_distance..s5af11s  )r%rmeanrzr{lenr|s rrd._EmbeddingDistanceChainMixin._hamming_distance s@ >>B7716? "5Q55A>>r$vectorscLURUR5n[5(aa[U[ 5R 5(a>U"USR SS5USR SS55R5nOU"USUS5n[U5$)zCompute the score based on the distance metric. Args: vectors (np.ndarray): The input vectors. Returns: float: The computed score. rr) rgrFr%rKrndarrayreshapeitemfloat)rVrr^rTs r_compute_score+_EmbeddingDistanceChainMixin._compute_scores!!$"6"67 >>j-/2I2IJJ71:--a4gaj6H6HB6OPUUWE71:wqz2EU|r$r#N)"r5r6r7r8r9r r,rDr __annotations__r.r:rFr dictstrrrPr model_configpropertylistrWr\rg staticmethodr`rarbrcrdrrr?r#r$rrArA[s#3EFJ F).7H7O7O)PO&P -$sCx.-T#s(^--^ $LT#Ydt "3*-C-C-C--*CsCsCsCC"5s5s5s55 5s5s5s55 ?S?S?S?? cer$rAc\rSrSrSr\S\4Sj5r\S\4Sj5r \S\ \4Sj5r SS\ \\ 4S \\S\ \\ 44S jjrSS\ \\ 4S \\S\ \\ 44S jjrSSSSS S .S\S\\S\S\\ \S\\ \\ 4S\S\ S\ 4SjjrSSSSS S .S\S\\S\S\\ \S\\ \\ 4S\S\ S\ 4SjjrSrg)EmbeddingDistanceEvalChaini+aUse embedding distances to score semantic difference between a prediction and reference. Examples: >>> chain = EmbeddingDistanceEvalChain() >>> result = chain.evaluate_strings(prediction="Hello", reference="Hi") >>> print(result) {'score': 0.5} rcg)zuReturn whether the chain requires a reference. Returns: bool: True if a reference is required, False otherwise. Tr#rUs rrequires_reference-EmbeddingDistanceEvalChain.requires_reference6sr$c6SURRS3$)N embedding_ _distancerFvaluerUs revaluation_name*EmbeddingDistanceEvalChain.evaluation_name?sD00667yAAr$c SS/$)MReturn the input keys of the chain. Returns: List[str]: The input keys. prediction referencer#rUs r input_keys%EmbeddingDistanceEvalChain.input_keysCsk**r$Ninputs run_managercURRUSUS/5n[5(a[5nUR U5nUR U5nSU0$)zCompute the score for a prediction and reference. Args: inputs (Dict[str, Any]): The input data. run_manager (Optional[CallbackManagerForChainRun], optional): The callback manager. Returns: Dict[str, Any]: The computed score. rrrTrDembed_documentsr%rarrayrrVrrrrrTs r_call EmbeddingDistanceEvalChain._callLsc//11 L !6+#6 7  >>Bhhw'G##G,r$c# URRUSUS/5IShvN n[5(a[5nUR U5nUR U5nSU0$NC7f)aAsynchronously compute the score for a prediction and reference. Args: inputs (Dict[str, Any]): The input data. run_manager (AsyncCallbackManagerForChainRun, optional): The callback manager. Returns: Dict[str, Any]: The computed score. rrNrTrDaembed_documentsr%rrrrs r_acall!EmbeddingDistanceEvalChain._acalldss88|${#    >>Bhhw'G##G, 'A/A-AA/F)r callbackstagsmetadatainclude_run_inforrrrrrkwargsc <U"XS.UUUUS9nURU5$)aEvaluate the embedding distance between a prediction and reference. Args: prediction (str): The output string from the first model. reference (str): The reference string (required) callbacks (Callbacks, optional): The callbacks to use. **kwargs (Any): Additional keyword arguments. Returns: dict: A dictionary containing: - score: The embedding distance between the two predictions. rrrrrrrr\ rVrrrrrrrrYs r_evaluate_strings,EmbeddingDistanceEvalChain._evaluate_stringss32",E-  ##F++r$cj# URXS.UUUUS9IShvN nURU5$N7f)aAsynchronously evaluate the embedding distance between a prediction and reference. Args: prediction (str): The output string from the first model. reference (str): The output string from the second model. callbacks (Callbacks, optional): The callbacks to use. **kwargs (Any): Additional keyword arguments. Returns: dict: A dictionary containing: - score: The embedding distance between the two predictions. rrNacallr\rs r_aevaluate_strings-EmbeddingDistanceEvalChain._aevaluate_stringssI2zz",E- "  ##F++ 313r#rq)r5r6r7r8r9rrrrrrrrrrr rrrrrrr?r#r$rrr+sDBBB+DI++=A S#X 89  c3h  6BF S#X =>  c3h  >$(#$(-1!& , ,C= ,  , tCy! ,4S>* , , ,  ,L$(#$(-1!& , ,C= ,  , tCy! ,4S>* , , ,  , ,r$rcb\rSrSrSr\S\\4Sj5r\S\4Sj5r SS\ \\ 4S\ \ S\ \\ 44S jjrSS\ \\ 4S\ \S\ \\ 44S jjrSSSS S .S \S\S\S\ \\S\ \ \\ 4S\S\ S\ 4SjjrSSSS S .S \S\S\S\ \\S\ \ \\ 4S\S\ S\ 4SjjrSrg)"PairwiseEmbeddingDistanceEvalChainizUse embedding distances to score semantic difference between two predictions. Examples: >>> chain = PairwiseEmbeddingDistanceEvalChain() >>> result = chain.evaluate_string_pairs(prediction="Hello", prediction_b="Hi") >>> print(result) {'score': 0.5} rc SS/$)rr prediction_br#rUs rr-PairwiseEmbeddingDistanceEvalChain.input_keyssn--r$c6SURRS3$)Npairwise_embedding_rrrUs rr2PairwiseEmbeddingDistanceEvalChain.evaluation_names$T%9%9%?%?$@ JJr$NrrcURRUSUS/5n[5(a[5nUR U5nUR U5nSU0$)zCompute the score for two predictions. Args: inputs (Dict[str, Any]): The input data. run_manager (CallbackManagerForChainRun, optional): The callback manager. Returns: Dict[str, Any]: The computed score. rrrTrrs rr(PairwiseEmbeddingDistanceEvalChain._callsf//11|$~&   >>Bhhw'G##G,r$c# URRUSUS/5IShvN n[5(a[5nUR U5nUR U5nSU0$NC7f)zAsynchronously compute the score for two predictions. Args: inputs (Dict[str, Any]): The input data. run_manager (AsyncCallbackManagerForChainRun, optional): The callback manager. Returns: Dict[str, Any]: The computed score. rrNrTrrs rr)PairwiseEmbeddingDistanceEvalChain._acallss88|$~&    >>Bhhw'G##G, rF)rrrrrrrrrrrc <U"XS.UUUUS9nURU5$)a Evaluate the embedding distance between two predictions. Args: prediction (str): The output string from the first model. prediction_b (str): The output string from the second model. callbacks (Callbacks, optional): The callbacks to use. tags (List[str], optional): Tags to apply to traces metadata (Dict[str, Any], optional): metadata to apply to **kwargs (Any): Additional keyword arguments. Returns: dict: A dictionary containing: - score: The embedding distance between the two predictions. rrrr rVrrrrrrrrYs r_evaluate_string_pairs9PairwiseEmbeddingDistanceEvalChain._evaluate_string_pairss34",K-  ##F++r$cj# URXS.UUUUS9IShvN nURU5$N7f)a7Asynchronously evaluate the embedding distance between two predictions. Args: prediction (str): The output string from the first model. prediction_b (str): The output string from the second model. callbacks (Callbacks, optional): The callbacks to use. tags (List[str], optional): Tags to apply to traces metadata (Dict[str, Any], optional): metadata to apply to traces **kwargs (Any): Additional keyword arguments. Returns: dict: A dictionary containing: - score: The embedding distance between the two predictions. rrNrrs r_aevaluate_string_pairs:PairwiseEmbeddingDistanceEvalChain._aevaluate_string_pairs7sI8zz",K- "  ##F++ rr#rq)r5r6r7r8r9rrrrrrrrr rrrrrrrr?r#r$rrrs.DI..KKK =A S#X 89  c3h  <BF S#X =>  c3h  @ $$(-1!&!,!, !,  !, tCy! !,4S>*!,!,!, !,P $$(-1!&#,#, #,  #, tCy! #,4S>*#,#,#, #,#,r$r)*r9 functoolsloggingenumr importlibrtypingrrlangchain_core.callbacksr langchain_core.callbacks.managerrr langchain_core.embeddingsr langchain_core.utilsr pydanticr r langchain.chains.baserlangchain.evaluation.schemarrlangchain.schemarr getLoggerr5r! lru_cacherr%r,rr.rArrr#r$rrsF .1)&'P$s   8 $ Q d   J0T$M5M`V,!=V,rV, V,r$