Ig1pddlZddlmZmZmZmZddlZddlmZddl m Z m Z ddl m Z mZGdde eZy)N)Dict GeneratorListOptional) Embeddings)get_from_dict_or_envpre_init) BaseModel ConfigDictc *eZdZUdZdZeed< dZeed< dZeed< dZ eed< dZ eed< iZ e ed< d Z eed < ed Zed edefdZdefdZdedefdZdeed edefdZ ddeed eedeeefdZdedeefdZy)SambaStudioEmbeddingsa3SambaNova embedding models. To use, you should have the environment variables ``SAMBASTUDIO_EMBEDDINGS_BASE_URL``, ``SAMBASTUDIO_EMBEDDINGS_BASE_URI`` ``SAMBASTUDIO_EMBEDDINGS_PROJECT_ID``, ``SAMBASTUDIO_EMBEDDINGS_ENDPOINT_ID``, ``SAMBASTUDIO_EMBEDDINGS_API_KEY`` set with your personal sambastudio variable or pass it as a named parameter to the constructor. Example: .. code-block:: python from langchain_community.embeddings import SambaStudioEmbeddings embeddings = SambaStudioEmbeddings(sambastudio_embeddings_base_url=base_url, sambastudio_embeddings_base_uri=base_uri, sambastudio_embeddings_project_id=project_id, sambastudio_embeddings_endpoint_id=endpoint_id, sambastudio_embeddings_api_key=api_key, batch_size=32) (or) embeddings = SambaStudioEmbeddings(batch_size=32) (or) # CoE example embeddings = SambaStudioEmbeddings( batch_size=1, model_kwargs={ 'select_expert':'e5-mistral-7b-instruct' } ) sambastudio_embeddings_base_urlsambastudio_embeddings_base_uri!sambastudio_embeddings_project_id"sambastudio_embeddings_endpoint_idsambastudio_embeddings_api_key model_kwargs batch_size)protected_namespacesvaluesreturnct|dd|d<t|ddd|d<t|dd|d<t|d d |d <t|d d |d <|S) z?Validate that api key and python package exists in environment.rSAMBASTUDIO_EMBEDDINGS_BASE_URLrSAMBASTUDIO_EMBEDDINGS_BASE_URIapi/predict/generic)defaultr!SAMBASTUDIO_EMBEDDINGS_PROJECT_IDr"SAMBASTUDIO_EMBEDDINGS_ENDPOINT_IDrSAMBASTUDIO_EMBEDDINGS_API_KEY)r)clsrs e/var/www/html/answerous/venv/lib/python3.12/site-packages/langchain_community/embeddings/sambanova.pyvalidate_environmentz*SambaStudioEmbeddings.validate_environmentEs5I 57X5 015I  - -) 5 01 7K  / /7 23 8L  0 08 34 4H 46V4 /0 c  d|jvr |j}nL|jjDcic]'\}}|t|jt |d)}}}t j|}|Scc}}w)z Get the tuning parameters to use when calling the model Returns: The tuning parameters as a JSON string. api/v2/predict/generic)typevalue)rritemsr)__name__strjsondumps)selftuning_params_dictkv tuning_paramss r$_get_tuning_paramsz(SambaStudioEmbeddings._get_tuning_params`s $t'K'K K!%!2!2 "..446"AqDG,,s1v>>" " #56  "s,A?pathc>|jd|jd|S)z Return the full API URL for a given path. :param str path: the sub-path :returns: the full API URL for the sub-path :rtype: str /)rr)r0r6s r$ _get_full_urlz#SambaStudioEmbeddings._get_full_urlqs,667q9]9]8^^_`d_effr&textsc#VKtdt||D] }||||zyw)afGenerator for creating batches in the embed documents method Args: texts (List[str]): list of strings to embed batch_size (int, optional): batch size to be used for the embedding model. Will depend on the RDU endpoint used. Yields: List[str]: list (batch) of strings of size batch size rN)rangelen)r0r:ris r$_iterate_over_batchesz+SambaStudioEmbeddings._iterate_over_batches{s5q#e*j1 ,AA N+ + ,s')Nc| |j}tj}|j|jd|j }t j|j}g}d|jvr|j||D]}||d}|j|d|ji|} | jdk7r%td| jd| j | j d } |j!| |Sd |jvr|j||D]}t%|D cgc] \} } d | | d } } } | |d}|j|d|ji|} | jdk7r%td| jd| j | j dD cgc]} | d } } |j!| |Sd|jvr|j||D]}||d}|j|d|ji|} | jdk7r%td| jd| j |j'dr| j d} n| j d} |j!| |St)d|jd#t"$rt#d | j wxYwcc} } wcc} w#t"$rt#d| j wxYw#t"$rt#d| j wxYw)a<Returns a list of embeddings for the given sentences. Args: texts (`List[str]`): List of texts to encode batch_size (`int`): Batch size for the encoding Returns: `List[np.ndarray]` or `List[tensor]`: List of embeddings for the given sentences r8api/predict/nlpinputsparamskeyheadersr.1Sambanova /complete call failed with status code . Details: data%'data' not found in endpoint responser(itemidr*r+rDr+r*&'items' not found in endpoint responser instancesrD select_expert predictions,'predictions' not found in endpoint responsehandling of endpoint uri:  not implemented)rrequestsSessionr9rrr.loadsr5rr?postr status_code RuntimeErrortextextendKeyError enumerateget ValueError)r0r:r http_sessionurlrD embeddingsbatchrKresponse embeddingr>rMr+s r$embed_documentsz%SambaStudioEmbeddings.embed_documentss  J'')   556a8_8_7` a D3356  D D D33E:F "'6:',,"D$G$GH- ''3.&K#//0 hmm_N (  7I%%i0 Xo&)M)M M33E:F ENuEU:A!TT!:5"'&9',,"D$G$GH- ''3.&K#//0 hmm_N;C==?7;S T4g TI T%%i0# l=#d&J&J J33E:F %*f=',,"D$G$GH- ''3.&K#//0 hmm_N zz/2$,MMOM$B $,MMOM$B %%i0# : ,T-Q-Q,RRbc s "? !U"@ 2 "F s=/$KK97L K?L,A L+$K6?L$L(+$Mr_cBtj}|j|jd|j}t j |j}d|jvrs|g|d}|j|d|ji|}|jdk7r%td|jd|j |j d d }|Sd |jvryd |dg|d}|j|d|ji|}|jdk7r%td|jd|j |j dd d}|Sd|jvr|g|d}|j|d|ji|}|jdk7r%td|jd|j |jdr|j dd }n|j dd }|St!d|jd#t$rtd |j wxYw#t$rtd|j wxYw#t$rtd|j wxYw)aReturns a list of embeddings for the given sentences. Args: sentences (`List[str]`): List of sentences to encode Returns: `List[np.ndarray]` or `List[tensor]`: List of embeddings for the given sentences r8rArBrErFrHrIrJrKrrLr(item0rNrPr+r*rQrrRrTrUrVrWrX)rYrZr9rrr.r[r5rr\rr]r^r_rarcrd)r0r_rerfrDrKrirjs r$ embed_queryz!SambaStudioEmbeddings.embed_querys  '')   556a8_8_7` a D3356  D D D#f7D#(( C CD)H ##s*"G++,M(--J $MMOF3A6 na&)M)M M%,t<=PD#(( C CD)H ##s*"G++,M(--J $MMOG4Q7@ F9#d&J&J J"&6:D#(( C CD)H ##s*"G++,M(--J ::o. (  >q AI (  >q AI ,T-Q-Q,RRbc e ;MMO ( <MMO . BMMO s$ H,I>I:,$I$I7:$J)N)r, __module__ __qualname____doc__rr-__annotations__rrrrrdictrintr model_configr rr%r5r9rrr?rfloatrkrnrr&r$r r s'!F,.#S-+-#S--/%s/-.0&0.*,"C,L$2J-26L $44C"g#g#g ,49 ,# ,) ,=Ab#Yb,4SMb d5k bHSSU Sr&r )r.typingrrrrrYlangchain_core.embeddingsrlangchain_core.utilsrr pydanticr r r rr&r$r{s* 220?*tIztr&