dh'SrSSKrSSKJr SSKJrJrJrJrJ r J r SSK r SSK r SSKrSSKJr SSKJr SSKJrJrJr S/r"S S\\5r"S S 5rg) z-written under MIT Licence, Michael Feil 2023.N)ThreadPoolExecutor)AnyCallableDictListOptionalTuple) Embeddings)get_from_dict_or_env) BaseModel ConfigDictmodel_validatorInfinityEmbeddingsc\rSrSr%Sr\\S'Sr\\S'Sr\ \S'\ "SS 9r \ "S S 9\ S \S \ 4Sj55rS\\S \\\4SjrS\\S \\\4SjrS\S \\4SjrS\S \\4SjrSrg)raSelf-hosted embedding models for `infinity` package. See https://github.com/michaelfeil/infinity This also works for text-embeddings-inference and other self-hosted openai-compatible servers. Infinity is a package to interact with Embedding Models on https://github.com/michaelfeil/infinity Example: .. code-block:: python from langchain_community.embeddings import InfinityEmbeddings InfinityEmbeddings( model="BAAI/bge-small", infinity_api_url="http://localhost:7997", ) modelzhttp://localhost:7997infinity_api_urlNclientforbid)extrabefore)modevaluesreturncD[USS5US'[USS9US'U$)z?Validate that api key and python package exists in environment.rINFINITY_API_URL)hostr)r &TinyAsyncOpenAIInfinityEmbeddingClient)clsrs _/var/www/html/shao/venv/lib/python3.13/site-packages/langchain_community/embeddings/infinity.pyvalidate_environment'InfinityEmbeddings.validate_environment3s? &: &(:& !"B*+ x textscNURRURUS9nU$)zCall out to Infinity's embedding endpoint. Args: texts: The list of texts to embed. Returns: List of embeddings, one for each text. rr$)rembedrselfr$ embeddingss r embed_documents"InfinityEmbeddings.embed_documentsAs/[[&&**' r#cj# URRURUS9IShvN nU$N7f)zAsync call out to Infinity's embedding endpoint. Args: texts: The list of texts to embed. Returns: List of embeddings, one for each text. r&N)raembedrr(s r aembed_documents#InfinityEmbeddings.aembed_documentsPs< ;;--**.    s (313textc,URU/5S$)zvCall out to Infinity's embedding endpoint. Args: text: The text to embed. Returns: Embeddings for the text. r)r+)r)r1s r embed_queryInfinityEmbeddings.embed_query_s##TF+A..r#cL# URU/5IShvN nUS$N 7f)z|Async call out to Infinity's embedding endpoint. Args: text: The text to embed. Returns: Embeddings for the text. Nr)r/)r)r1r*s r aembed_queryInfinityEmbeddings.aembed_queryjs+ 00$88 !}9s $" $)__name__ __module__ __qualname____firstlineno____doc__str__annotations__rrrr model_configr classmethodrr!rfloatr+r/r3r6__static_attributes__r8r#r rrs& J#3c3FCL(# $ 3 $  T#Y 4U 3D  DI $tE{:K  / /U / s tE{ r#c \rSrSrSrSS\S\\RSS4Sjjr \ \ 4S\ \S \ S\\ \\ 44S jj5rS\ \S\ \ \4S jr\ S \ \ \S\ \4S j5rS\S\ \S\\\44SjrS\S\ \S\ \ \4SjrS\S\ \S\ \ \4SjrS\RS\\\4S\ \ \4SjrS\S\ \S\ \ \4SjrSrg)rwaHelper tool to embed Infinity. It is not a part of Langchain's stable API, direct use discouraged. Example: .. code-block:: python mini_client = TinyAsyncInfinityEmbeddingClient( ) embeds = mini_client.embed( model="BAAI/bge-small", text=["doc1", "doc2"] ) # or embeds = await mini_client.aembed( model="BAAI/bge-small", text=["doc1", "doc2"] ) Nr aiosessionrcXlX lURb[UR5S:a [S5eSUlg)Nz( param `host` must be set to a valid url)rrFlen ValueError _batch_size)r)rrFs r __init__/TinyAsyncOpenAIInfinityEmbeddingClient.__init__s;  $ 99 DII 2GH Hr#r$sorterc^[U5S:XaUS4$[R"UVs/sH o!"U5*PM sn5mTVs/sHo0UPM nnUU4Sj4$s snfs snf)aSort texts in ascending order, and delivers a lambda expr, which can sort a same length list https://github.com/UKPLab/sentence-transformers/blob/ c5f93f70eca933c78695c5bc686ceda59651ae3b/sentence_transformers/SentenceTransformer.py#L156 Args: texts (List[str]): _description_ sorter (Callable, optional): _description_. Defaults to len. Returns: Tuple[List[str], Callable]: _description_ Example: ``` texts = ["one","three","four"] perm_texts, undo = self._permute(texts) texts == undo(perm_texts) ``` cU$Nr8)ts r ATinyAsyncOpenAIInfinityEmbeddingClient._permute..sAr#c^>[R"T5Vs/sHoUPM sn$s snfrS)npargsort)unsorted_embeddingsidxlength_sorted_idxs r rUrVs-02 ;L0M: 0M $0M: : s *)rJrXrY)r$rOsenr[ texts_sortedr\s @r _permute/TinyAsyncOpenAIInfinityEmbeddingClient._permutesr0 u:?+% %JJ'F 'FG.?@.?sc .? @   (G@s A A"c[U5S:XaU/$/n[S[U5UR5H"nURXX0R-5 M$ U$)a splits Lists of text parts into batches of size max `self._batch_size` When encoding vector database, Args: texts (List[str]): List of sentences self._batch_size (int, optional): max batch size of one request. Returns: List[List[str]]: Batches of List of sentences rQr)rJrangerLappend)r)r$batches start_indexs r _batch-TinyAsyncOpenAIInfinityEmbeddingClient._batchsX u:?7N CJ0@0@AK NN5{=M=M/MN OBr#batch_of_textsc[U5S:Xa[US5S:XaUS$/nUHnURU5 M U$)NrQr)rJextend)rhr$sublists r _unbatch/TinyAsyncOpenAIInfinityEmbeddingClient._unbatchsK ~ ! #N1,=(>!(C!!$ $%G LL !& r#rc F[URS3SS0[UUS9S9$)zBuild the kwargs for the Post request, used by sync Args: model (str): _description_ texts (List[str]): _description_ Returns: Dict[str, Collection[str]]: _description_ z /embeddingsz content-typezapplication/json)inputr)urlheadersjson)dictr)r)rr$s r _kwargs_post_request;TinyAsyncOpenAIInfinityEmbeddingClient._kwargs_post_requests<99+[) 2  r# batch_textsc [R"S0URXS9D6nURS:wa%[ SURSUR 35eUR 5SVs/sHoDSPM sn$s snf)Nr&5Infinity returned an unexpected response with status : data embeddingr8)requestspostrt status_code Exceptionr1rr)r)rrvresponsees r _sync_request_embed:TinyAsyncOpenAIInfinityEmbeddingClient._sync_request_embeds== ''e'G    3 &G''(8==/; )1 (?@(?1+(?@@@s. A>ctURU5up4URU5nURU/[U5-U4n[U5S:Xa[ [ U65nO,[ S5n[ UR "U65nSSS5 URW5n U"U 5n U $!,(df  N)=f)zcall the embedding of model Args: model (str): to embedding model texts (List[str]): List of sentences to embed. Returns: List[List[float]]: List of vectors for each sentence rQ N)r_rfrrJlistmaprrl) r)rr$ perm_textsunpermute_funcperm_texts_batchedmap_argsembeddings_batch_permpembeddings_permr*s r r',TinyAsyncOpenAIInfinityEmbeddingClient.embeds&*]]5%9" ![[4  $ $ Gc,- -   ! "a '$(h$8 !#B'1(,QUUH-=(>%(--(=>#O4  ('s -B)) B7sessionkwargsct# UR"S0UD6IShvN nURS:wa%[SURSUR35eUR 5IShvN SnUVs/sHoUSPM snsSSS5IShvN $NyN.s snfN!,IShvN (df  g=f7f)Nrxryrzr{r|r8)r~statusrr1rr)r)rrrr|rs r _async_request5TinyAsyncOpenAIInfinityEmbeddingClient._async_requests<<)&))X#%K'r(--: (}}.7I,56IqkNI6*)) /6*)))siB8BB8A B&B' B2 B?B B8 BB8BBB8B5$B' % B51B8c# URU5up4URU5n[R"S[R"SS9S9IShvN n[ R "UVs/sH nURUURXS9S9PM" sn6IShvN nSSS5IShvN URW5n U"U 5n U $Nus snfN6N(!,IShvN (df  N==f7f)zcall the embedding of model, async method Args: model (str): to embedding model texts (List[str]): List of sentences to embed. Returns: List[List[float]]: List of vectors for each sentence Tr)limit) trust_env connectorNr&)rr) r_rfaiohttp ClientSession TCPConnectorasynciogatherrrtrl) r)rr$rrrrrTrrr*s r r.-TinyAsyncOpenAIInfinityEmbeddingClient.aembed's&*]]5%9" ![[4((g&:&:&D  *1.. 0  0 '' '#88u8N(0 +% !  --(=>#O4  %    sfAC-CC-C+&C CCC C-(C) C- CC-C*C  C*&C-)rLrFr)zhttp://localhost:7797/v1N)r9r:r;r<r=r>rrrrM staticmethodrJrrr r_rfrrlrrtrBrr'rr.rCr8r#r rrws2/6:  W223   -0 Cy "* tCy(" #  BDI$tCy/(d3iT#Y # d3i DcN , A A'+Cy A d5k  A3tCyT$u+5F: 7,, 76:38n 7 d5k  7#d3iDersN31==05;;  cJcLMMr#