kQjyB|ddlZddlZddlmZmZmZmZmZmZm Z ddl m Z ddl m Z ddlmZddlZddlZddlmZdd lmZdd lmZdd lmZmZmZerddlZd d ddd d dddddddd ZdZ dZ!hdZ"de#de$fdZ%de de$fdZ&dee#e'e ffdZ(deeefdeefdZ)dedee#fdZ*edGdd eZ+y)!N)ClassVar TYPE_CHECKINGListUnionAny GeneratorOptional)Path)urlparse)BytesIO)attempt_import_or_raise)EmbeddingFunction)register)api_key_not_found_helpIMAGESTEXTii@Bi}) voyage-4 voyage-4-litevoyage-4-largevoyage-context-3voyage-3.5-lite voyage-3.5 voyage-3-litevoyage-3voyage-multimodal-3voyage-finance-2voyage-multilingual-2 voyage-law-2 voyage-code-2ic t|}t|jxrt|jS#t$rYywxYw)NF)r boolschemenetloc Exception)textparseds `/home/oem/CyberBrain26-unified/.venv/lib/python3.12/site-packages/lancedb/embeddings/voyageai.py is_valid_urlr++s?$FMM":tFMM':: s69 AA>.avi.gif.m4v.mkv.mov.mp4.webmurlreturnc~t|}|jjtfdtDS)z7Check if URL points to a video file based on extension.c3@K|]}j|ywN)endswith).0extpaths r* zis_video_url..:s>ct}}S!>s)r r;loweranyVIDEO_EXTENSIONS)r3r)r;s @r* is_video_urlr@6s0 c]F ;;   D >-=> >>r;cB|jjtvS)z6Check if file path is a video file based on extension.)suffixr=r?)r;s r* is_video_pathrD=s ;;   "2 22rA input_datactdd}t|tr+t|rt |rd|d}n d|d}nd|d}nt||j rht }|jd j|d tj|jjd }d d|zd}n~t|tr|jt |}t }|jd j|d tj|jjd }d d|zd}nt|trt!|rOt|d5}|j#}dddtjjd }d|d}n|j|}t }|jd j|d tj|jjd }d d|zd}n t%dd|giS#1swYxYw)N PIL.Imagepillow video_url)typerI image_url)rJrKr()rJr(RGBJPEG)formatzutf-8 image_base64zdata:image/jpeg;base64,)rJrOrb video_base64)rJrQ6Each input should be either str, bytes, Path or Image.content)r isinstancestrr+r@Imager convertsavebase64 b64encodegetvaluedecodebytesopenr rDread ValueError) rE PIL_ImagerSbufferedimg_strimgf video_bytes video_strs r*transform_inputrhBs' X>I*c"  #J'#.ZH#.ZH%z:G J  095!&&x&?""8#4#4#67>>wG"5?  J &nnWZ019 E8""8#4#4#67>>wG"5?  J %  $j$' '1ffh  '((5<I  Iinputsctddt|tttj fr|g}nt|t rn|t|tjr|j}nQt|tjr|jj}ntdt|dtfd|Ds td|Dcgc] }t|c}Scc}w)7 Sanitize the input to the embedding function. rGrH Input type z# not allowed with multimodal model.c3jK|]*}t|tttjf,ywr7)rTrUr]r rV)r9xras r*r<z,sanitize_multimodal_input..s%Raz!c5$ @ARs03rR)rrTrUr]r rVlistpaArray to_pylist ChunkedArraycombine_chunksr`rJallrh)riiras @r*sanitize_multimodal_inputrwvs( X>I&3tY__=> FD ! FBHH %!!# FBOO ,&&(224$v,'J K   R6R RQRR(. /1OA  // /s(C=ct|tr|g}nt|ttjfr t|}n|t|t j r|j}nQt|t jr|jj}ntdt|dtd|Ds td|S)rkrlz not allowed with text model.c3<K|]}t|tywr7)rTrU)r9rns r*r<z&sanitize_text_input..s2az!S!2szEach input should be str.) rTrUronpndarrayrprqrrrsrtr`rJru)ris r*sanitize_text_inputr|s&# FT2::. /f FBHH %!!# FBOO ,&&(224;tF|n4QRSS 262 2455 MrAvoyageaiceZdZUdZeed<dZeeed<dZ e ed<dgZ e e ed<gdZ e e ed <gd Ze ed <d dgZe ed <dgZe ed<defdZdefdZdZdZdeedfdeej0fdZdeeefdeej8fdZdeedeeeddffdZ d!dede fdZ! d!deededeej8fdZ"e#d Z$y)"VoyageAIEmbeddingFunctionad An embedding function that uses the VoyageAI API https://docs.voyageai.com/docs/embeddings Parameters ---------- name : str The name of the model to use. List of acceptable models: * voyage-4 (1024 dims, general-purpose and multilingual retrieval) * voyage-4-lite (1024 dims, optimized for latency and cost) * voyage-4-large (1024 dims, best retrieval quality) * voyage-context-3 * voyage-3.5 * voyage-3.5-lite * voyage-3 * voyage-3-lite * voyage-multimodal-3 * voyage-multimodal-3.5 * voyage-finance-2 * voyage-multilingual-2 * voyage-law-2 * voyage-code-2 output_dimension : int, optional The output dimension for models that support flexible dimensions. Currently only voyage-multimodal-3.5 supports this feature. Valid options: 256, 512, 1024 (default), 2048. Examples -------- import lancedb from lancedb.pydantic import LanceModel, Vector from lancedb.embeddings import EmbeddingFunctionRegistry voyageai = EmbeddingFunctionRegistry .get_instance() .get("voyageai") .create(name="voyage-3") class TextModel(LanceModel): text: str = voyageai.SourceField() vector: Vector(voyageai.ndims()) = voyageai.VectorField() data = [ { "text": "hello world" }, { "text": "goodbye world" }] db = lancedb.connect("~/.lancedb") tbl = db.create_table("test", schema=TextModel, mode="overwrite") tbl.add(data) nameNoutput_dimensionclientzvoyage-multimodal-3.5_FLEXIBLE_DIM_MODELS)i_VALID_DIMENSIONS) rrrrrrrrr r!r"text_embedding_modelsrmultimodal_embedding_modelsrcontextual_embedding_models model_namec*||jvxsd|vS)N multimodal)rselfrs r*_is_multimodal_modelz.VoyageAIEmbeddingFunction._is_multimodal_models $:: : Xlj>X rAc*||jvxsd|vS)Ncontext)rrs r*_is_contextual_modelz.VoyageAIEmbeddingFunction._is_contextual_modelsT===XjAXXrAc|j|jvrc|jV|j|jvr2t d|jd|jd|j|jSy|jdk(ry|jdk(ry|jd vryt d |jd ) NzInvalid output_dimension z for z. Valid options: rrrr"i) rrrrrrrrrr r!zModel z not supported)rrrrr`)rs r*ndimszVoyageAIEmbeddingFunction.ndimss 9911 1$$0((0F0FF$3D4I4I3JK#yyk):4;Q;Q:RT,,, 99 ' YY/ ) YY   vdii[?@ @rAc l|j|jvr|j|j|d<|S)zHGet kwargs for multimodal embed call, including output_dimension if set.r)rrr)rkwargss r*_get_multimodal_kwargsz0VoyageAIEmbeddingFunction._get_multimodal_kwargss5 9911 1d6K6K6W)-)>)>F% & rAqueryzPIL.Image.Imager4ctj}|j|jr5|jdi|}|j d|gg|jdd|}nn|j |jr2|jd|gg|jdd|}|jd}n!|jd|g|jdd|}|jdgS)a. Compute the embeddings for a given user query Parameters ---------- query : Union[str, PIL.Image.Image] The query to embed. A query can be either text or an image. Returns ------- List[np.array]: the list of embeddings rrimodel input_typertextsrr) r _get_clientrrrmultimodal_embedrcontextualized_embedresultsembed embeddings)rrargsrrresults r*compute_query_embeddingsz2VoyageAIEmbeddingFunction.compute_query_embeddings s+668  $ $TYY /0T00:6:F,V,,y gIOF & &tyy 10V00y gIOF^^A&F!V\\gTYY7FLF!!!$%%rArictj}|j|jrvt |}t d|D}|r>|j di|}|jd||jdd|}|jS|Dcgc] }|ddd}}n t|}|j||fddi|Scc}w) aX Compute the embeddings for the inputs Parameters ---------- inputs : Union[TEXT, IMAGES] The inputs to embed. The input can be either str, bytes, Path (to an image), PIL.Image or list of these. Returns ------- List[np.array]: the list of embeddings c3PK|]}|ddjddk7 yw)rSrrJr(N)get)r9inps r*r<zFVoyageAIEmbeddingFunction.compute_source_embeddings..Vs.rrrr|_embed_with_batching) rrirrr sanitized has_imagesrrs r*compute_source_embeddingsz3VoyageAIEmbeddingFunction.compute_source_embeddingsAs +668  $ $TYY /1&9ICLJ444>v>000$DII*PV(((;DECc)nQ'/EFE(0F)t(( F '1 5;  FsCrc#K|sytj|jd}g}d}|j||j}|Dcgc] }t |}}t |D]D\} } || } |r"t |t k\s|| z|kDr|g}d}|j| || z }F|r|yycc}ww)aL Generate batches of texts based on token limits using a generator. Parameters ---------- client : voyageai.Client The VoyageAI client instance. texts : List[str] List of texts to batch. Yields ------ List[str]: Batches of texts. Nrr)r)VOYAGE_TOTAL_TOKEN_LIMITSrrtokenizelen enumerate BATCH_SIZEappend) rrrmax_tokens_per_batch current_batchcurrent_batch_tokens token_lists token_list token_countsrvr(n_tokenss r*_build_batchesz(VoyageAIEmbeddingFunction._build_batchesjs" 8< B9A!B>rc jjrAjdidttdtt j ffd }|Sjjr/dttdtt j ffd }|Sdttdtt j ffd }|S)a Get the appropriate embedding function based on model type. Parameters ---------- client : voyageai.Client The VoyageAI client instance. input_type : str Either "query" or "document" **kwargs Additional arguments to pass to the embedding API Returns ------- callable: A function that takes a batch of texts and returns embeddings. batchr4crt|}jd|jd}|jS)Nrr)rwrrr)r batch_inputsrrrmultimodal_kwargsrs r* embed_batchzBVoyageAIEmbeddingFunction._get_embed_function..embed_batchsI8? 000')))(  (((rAcxjd|gjd}|jdjS)Nrrr)rrrrrrrrrrs r*rzBVoyageAIEmbeddingFunction._get_embed_function..embed_batchsF444!7$)) NT~~a(333rAc\jd|jd}|jS)Nrr)rrrrs r*rzBVoyageAIEmbeddingFunction._get_embed_function..embed_batchs9%tyyZKQ(((rAr)rrrrrUrzarrayr)rrrrrrs```` @r*_get_embed_functionz-VoyageAIEmbeddingFunction._get_embed_functions&  $ $TYY / ; ; ; Ef E  )49 )bhh ) )   & &tyy 1 449 4bhh 4 4   )49 )bhh ) )  rAc |sgS|j|fd|i|}g}|j||D]}||}|j||S)a Embed texts with automatic batching based on token limits. Parameters ---------- client : voyageai.Client The VoyageAI client instance. texts : List[str] List of texts to embed. input_type : str Either "query" or "document" **kwargs Additional arguments to pass to the embedding API Returns ------- List[np.array]: List of embeddings. r)rrextend) rrrrrembed_fnall_embeddingsrbatch_embeddingss r*rz.VoyageAIEmbeddingFunction._embed_with_batchingsn*I,4++FTzTVT((7 4E'   ! !"2 3 4rActj`td}tjj d t d|jtjdt_tjS)Nr}VOYAGE_API_KEY)rrrosenvironrrClient)r}s r*rz%VoyageAIEmbeddingFunction._get_clientsa $ + + 3.z:Hzz~~./7&z2/7 +,0 % ,)///rA)r)%__name__ __module__ __qualname____doc__rU__annotations__rr intrrrrorrrrrrrrrrrzr{rrrrrrrcallablerr staticmethodrrrAr*rrsy5n I&*hsm*FH,C+D(4.D(>x~> #4 *?@W(XX);(<< s YsYAB &3 112& bjj &B' D&L)' bhh' R. !#Y. 49dD( ). b)34"%4 4n;E!!#Y!47! bhh!F00rAr),rYrtypingrrrrrrr pathlibr urllib.parser ior numpyrzpyarrowrputilrbaserregistryrutilsrrrPILrrr+r?rUr$r@rDr]rhrwr|rrrAr*rs QQQ!*#77 "$" M?c?d?333 1"c5$&6 71"h0eD&L&90d3i00c* *X0 1X0X0rA